MOOC and open courseware

www.aalto.fi

T 61.3025 Principles of Pattern Recognition Weekly Exercises with Solutions (in English), Aalto University, Finland, 2015

CSE E5430 Scalable Cloud Computing Weekly Exercises with Solutions (in English), Aalto University, Finland, Fall 2014

TO EXPLORE, not to be missed!

29 Dec 2014

Pattern Recognition Class, Universität Heidelberg, 2012 (Videos in English)

Syllabus:

1. Introduction

1.1 Applications of Pattern Recognition

1.2 k Nearest Neighbors Classification

1.3 Probability Theory

1.4 Statistical Decision Theory

2. Correlation Measures, Gaussian Models

2.1 Pearson Correlation

2.2 Alternative Correlation Measures

2.3 Gaussian Graphical Models

2.4 Discriminant Analysis

3. Dimensionality Reduction

3.1 Regularized LDA/QDA

3.2 Principal Component Analysis (PCA)

3.3 Bilinear Decompositions

4. Neural Networks

4.1 History of Neural Networks

4.2 Perceptrons

4.3 Multilayer Perceptrons

4.4 The Projection Trick

4.5 Radial Basis Function Networks

5. Support Vector Machines

5.1 Loss Functions

5.2 Linear Soft Margin SVM

5.3 Nonlinear SVM

6. Kernels, Random Forest

6.1 Kernels

6.2 One Class SVM

6.3 Random Forest

6.4 Random Forest Feature Importance

7. Regression

7.1 Least Squares Regression

7.2 Optimum Experimental Design

7.3 Case Study: Functional MRI

7.4 Case Study: Computer Tomography

7.5 Regularized Regression

8. Gaussian Processes

8.1 Gaussian Process Regression

8.2 GP Regression: Interpretation

8.3 Gaussian Stochastic Processes

8.4 Covariance Function

9. Unsupervised Learning

9.1 Kernel Density Estimation

9.2 Cluster Analysis

9.3 Expectation Maximization

9.4 Gaussian Mixture Models

10. Directed Graphical Models

10.1 Bayesian Networks

10.2 Variable Elimination

10.3 Message Passing

10.4 State Space Models

11. Optimization

11.1 The Lagrangian Method

11.2 Constraint Qualifications

11.3 Linear Programming

11.4 The Simplex Algorithm

12. Structured Learning

12.1 structSVM

12.2 Cutting Planes

26 Dec 2014

Convex Optimisation, Fall 2013, by Barnabas Poczos and Ryan Tibshirani, CMU

Overview and objectives

Nearly every problem in machine learning and statistics can be formulated in terms of the optimization of some function, possibly under some set of constraints. As we obviously cannot solve every problem in machine learning or statistics, this means that we cannot generically solve every optimization problem (at least not efficiently). Fortunately, many problems of interest in statistics and machine learning can be posed as optimization tasks that have special properties—such as convexity, smoothness, separability, sparsity etc.— permitting standardized, efficient solution techniques.

This course is designed to give a graduate level student a thorough grounding in these properties and their role in optimization, and a broad comprehension of algorithms tailored to exploit such properties. The main focus will be on convex optimization problems, though we will also discuss nonconvex problems at the end. We will visit and revisit important applications in statistics and machine learning. Upon completing the course, students should be able to approach an optimization problem (often derived from a statistics or machine learning context) and:

(1) identify key properties such as convexity, smoothness, sparsity, etc., and/or possibly reformulate the problem so that it possesses such desirable properties;

(2) select an algorithm for this optimization problem, with an understanding of the ad vantages and disadvantages of applying one method over another, given the problem and properties at hand;

(3) implement this algorithm or use existing software to efficiently compute the solution.

www.stat.cmu.edu

Machine Learning, Spring 2011, by Tom Mitchell, CMU

Machine Learning is concerned with computer programs that automatically improve their performance through experience (e.g., programs that learn to recognize human faces, recommend music and movies, and drive autonomous robots). This course covers the theory and practical algorithms for machine learning from a variety of perspectives. We cover topics such as Bayesian networks, decision tree learning, Support Vector Machines, statistical learning methods, unsupervised learning and reinforcement learning. The course covers theoretical concepts such as inductive bias, the PAC learning framework, Bayesian learning methods, margin based learning, and Occam's Razor. Short programming assignments include hands on experiments with various learning algorithms, and a larger course project gives students a chance to dig into an area of their choice. This course is designed to give a graduate level student a thorough grounding in the methodologies, technologies, mathematics and algorithms currently needed by people who do research in machine learning.

www.cs.cmu.edu

Homework with solutions

www.cs.cmu.edu

23dec 2014

Cambridge Machine Learning Slides, Spring 2014

LECTURE SYLLABUS

This year, the exposition of the material will be centered around three specific machine learning areas: 1) supervised non paramtric probabilistic inference using Gaussian processes, 2) the TrueSkill ranking system and 3) the latent Dirichlet Allocation model for unsupervised learning in text.

mlg.eng.cam.ac.uk

NPTEL, National Programme on Technology Enhanced Learning, India

NPTEL provides E learning through online Web and Video courses in Engineering, Science and humanities streams. The mission of NPTEL is to enhance the quality of Engineering education in the country by providing free online courseware.

nptel.ac.in

Probability Theory and Applications

nptel.ac.in

Pattern Recognition

nptel.ac.in

07 Dec_2014

Prof. Jürgen Schmidhuber's Home Page (Great resources! Not to be missed!)

Prof. Jürgen Schmidhuber's Artificial Intelligence team has won nine international competitions in machine learning and pattern recognition (more than any other AI research group) and seven independent best paper/best video awards, achieved the world's first superhuman visual classification results, Deep Learning since 1991, Winning Contests in Pattern Recognition and Sequence Learning Through Fast & Deep / Recurrent Neural Networks has pioneered Deep Learning methods for Artificial Neural Networks since 1991, and established the field of mathematically rigorous universal AI and optimal universal problem solvers. His formal theory of creativity & curiosity & fun explains art, science, music, and humor. He generalized algorithmic information theory, and the many worlds theory of physics, to obtain a minimal theory of all constructively computable universes, an elegant algorithmic theory of everything. Google & Apple and many other leading companies are now using the machine learning techniques developed in his group at the Swiss AI Lab IDSIA & USI & SUPSI (ex TUM CogBotLab). Since age 15 or so his main scientific ambition has been to build an optimal scientist through self improving AI, then retire. Progress is accelerating, are 40,000 years of human dominated history about to converge within the next few decades?

people.idsia.ch

ESAC DATA ANALYSIS AND STATISTICS WORKSHOP 2014

Model Fitting and Model Selection, Data Mining and Machine Learning, etc

ABOUT THE ESAC FACULTY

The ESAC Faculty was created in 2006 in order to foster an effective scientific environment at ESAC, and to to present a united face to the scientific work done at the centre. The faculty includes all active (i.e. publishing papers) research scientists at ESAC: ESA staff, Research Fellows, Science Contractors, and LAEFF members. For an insight into the founding principles, see the Overview of the ESAC Faculty presentation given at the first assembly.

The ESAC Faculty's main purpose is to stimulate and promote science activities at ESAC. For this it maintains an active and attractive visitor programme for short to medium term collaborative stays at ESAC, covering established researchers as well as young post docs, PhD and graduate students. The Faculty also supports visiting seminar speakers, conferences, workshops and travel not possibly via normal mission budgets.

ESAC Faculty members pursue their own research (as per the scientific interests of individual members), but are also involved in numerous internal and external collaborations (overview of Faculty Science at ESAC). Faculty members are also strongly involved in the ESAC Trainee programme.

www.cosmos.esa.int

04 Dec 2014

Richard Socher Deep Learning Tutorial (without Magic) for NLP, University of Montreal, LISA, ACL 2012

Richard Socher is a PhD student at Stanford working with Chris Manning and Andrew Ng. His research interests are machine learning for NLP and vision. He is interested in techniques that learn semantic features, capture recursive structure in multiple modalities and perform well across multiple supervised tasks. Most recently he developed several recursive deep learning models for compositionality in vector spaces, parsing, sentiment analysis, paraphrasing and word relation classification. In 2011, he was awarded the Yahoo! Key Scientific Challenges Program Award, the Distinguished Application Paper Award at ICML and a Microsoft Research Fellowship.

OUTLINE

PART I: The Basics

• Motivation

• From logistic regression to neural networks

• Theory: Backpropagation training

• Applications: Word vector learning, POS, NER

• Unsupervised pre training, multi task learning, and learning relations

PART II: Recursive Neural Networks

• Motivation

• Definition of RNNs

• Theory: Backpropagation through structure

• Applications: Sentiment Analysis, Paraphrase detection, Relation Classification

PART III: Applications and Discussion

• Overview of various NLP applications,

• Efficient reconstruction or prediction of high dimensional sparse vectors

• Discussion of future directions, advantages and limitations

www.acl2012.org

Introduction to Big Data with Apache Spark, EdX, 23 Feb 2015

Learn how to apply data science techniques using parallel programming in Apache Spark to explore big (and small) data.

About this Course

Organizations use their data for decision support and to build data intensive products and services, such as recommendation, prediction, and diagnostic systems. The collection of skills required by organizations to support these functions has been grouped under the term Data Science. This course will attempt to articulate the expected output of Data Scientists and then teach students how to use PySpark (part of Apache Spark) to deliver against these expectations. The course assignments include Log Mining, Textual Entity Recognition, Collaborative Filtering exercises that teach students how to manipulate data sets using parallel processing with PySpark.

This course covers advanced undergraduate level material. It requires a programming background and experience with Python (or the ability to learn it quickly). All exercises will use PySpark (part of Apache Spark), but previous experience with Spark or distributed computing is NOT required. Students should take this Python mini quiz before the course and take this Python mini course if they need to learn Python or refresh their Python knowledge.

22 Nov 2014

MLSS Iceland 2014

The Machine Learning Summer School will take place at Reykjavik University in Reykjavik, Iceland, from April 25 to May 4, 2014.

The field of machine learning is at the intersection of computer science, statistics, mathematics, and optimization. The Machine Learning Summer School (MLSS) is a great venue for graduate students, researchers, and professionals to learn about fundamental and advanced methods of machine learning, data analysis, and inference, from theory to practice.

The Machine Learning Summer School in Reykjavik will feature an exciting program with talks from leading experts in the field.

mlss2014.hiit.fi

Hugo Larochelle Neural Networks Lectures, University of Sherbrooke, 2013

(We've posted the French Lectures a long time ago but we've forgotten the English ones, we're fixing it now. Hugo Larochelle is a great teacher, his lectures break down complex concepts into small understandable ideas, not to be missed!)

These are the videos I use to teach my Neural networks class at Université de Sherbrooke. The videos, along with the slides and research paper references, are available here:

tinyurl.com

CIS 520 Machine Learning Course, Lyle Ungar, University of Pennsylvania

Course Description

CIS 520 provides a fundamental introduction to the mathematics, algorithms and practice of machine learning. Topics covered include:

Supervised learning: least squares regression, logistic regression, perceptron, naive Bayes, support vector machines. Model and feature selection, ensemble methods, boosting. Learning theory: Bias/variance tradeoff. Online learning.

Unsupervised learning: Clustering. K means. EM. Mixture of Gaussians. PCA.

Graphical models: HMMs, Bayesian and Markov networks. Inference. Variable elimination.

Audience

The course is aimed broadly at advanced undergraduates and beginning graduate students in computer science, electrical engineering, mathematics, physics, and statistics. Undergraduates who meet the prerequisites are particularly encouraged to enroll, as are students from other departments. This is a hard course; A good alternative for those with less linear algebra or time is CIS419/519 or, if you want a really nice, much easier intro, take the Coursera ML course. If unsure which to take, see this.

Reading Materials

Semi Optional Text: C. Bishop, Pattern Recognition and Machine Learning. 2007

Selected readings from other books and papers will be available on this web site.

Visualization Lab Course Wiki, Computer Science Division, University of California, Berkeley

vis.berkeley.edu

11 Nov 2014

10 years of Homeworks with Solutions and Lecture Slides, not to be missed !

Foundations of Machine Learning by Mehryar Mohri

Course Description

This course introduces the fundamental concepts and methods of machine learning, including the description and analysis of several modern algorithms, their theoretical basis, and the illustration of their applications. Many of the algorithms described have been successfully used in text and speech processing, bioinformatics, and other areas in real world products and services. The main topics covered are:

Probability tools, concentration inequalities

PAC model

Rademacher complexity, growth function, VC dimension

Perceptron, Winnow

Support vector machines (SVMs)

Kernel methods

Decision trees

Boosting

Density estimation, maximum entropy models

Logistic regression

Regression problems and algorithms

Ranking problems and algorithms

Halving algorithm, weighted majority algorithm, mistake bounds

Learning automata and transducers

Reinforcement learning, Markov decision processes (MDPs)

www.cs.nyu.edu

The Open Source Data Science Masters by Clare Corthell

The open source Data Science Masters

The open source curriculum for learning Data Science. Foundational in both theory and technologies, the OSDSM breaks down the core competencies necessary to make data useful.

TheInternet is Your Oyster

With Coursera, ebooks, Stack Overflow, and GitHub all free and open how can you afford not to take advantage of an open source education?

The Motivation

We need more Data Scientists.

...by 2018 the United States will experience a shortage of 190,000 skilled data scientists, and 1.5 million managers and analysts capable of reaping actionable insights from the big data deluge.

McKinsey Report Highlights the Impending Data Scientist Shortage 23 July 2013

There are little to no Data Scientists with 5 years experience, because the job simply did not exist.

David Hardtke How To Hire A Data Scientist 13 Nov 2012

An Academic Shortfall

Classic academic conduits aren't providing Data Scientists this talent gap will be closed differently.

Academic credentials are important but not necessary for high quality data science. The core aptitudes, curiosity, intellectual agility, statistical fluency, research stamina, scientific rigor, skeptical nature, that distinguish the best data scientists are widely distributed throughout the population.

We’re likely to see more uncredentialed, inexperienced individuals try their hands at data science, bootstrapping their skills on the open source ecosystem and using the diversity of modeling tools available. Just as data science platforms and tools are proliferating through the magic of open source, big data’s data scientist pool will as well.

And there’s yet another trend that will alleviate any talent gap: the democratization of data science. While I agree wholeheartedly with Raden’s statement that “the crème de la crème of data scientists will fill roles in academia, technology vendors, Wall Street, research and government,” I think he’s understating the extent to which autodidacts, the self taught, uncredentialed, data passionate people, will come to play a significant role in many organizations’ data science initiatives.

James Kobielus, Closing the Talent Gap 17 Jan 2013

Ready?

datasciencemasters.org

github.com

Open Machine Learning Workshop organized by Alekh Agarwal, Alina Beygelzimer, and John Langford, August 2014

The goal of this workshop is to inform people about open source machine learning systems being developed, aid the coordination of such projects, and discuss future plans.

hunch.net

Deep Learning: Machine Perception and Its Applications by Adam Gibson (a practical perspective) , Hakka Labs, 07 11 2014

Live Q&A on Machine Learning and IoT for Developers by Steve Teixeira, Raymond Laghaeian, Seth Juarez, TechEd Europe 2014

09 11 2014

Deep Learning for Vision & Applications of deep learning to temporal data by Graham Taylor from University of Guelph (Canada), CIMAT, 2014

www.uoguelph.ca

Maestria y Doctorado en Ciencias de la Computación de CIMAT Youtube Channel, Mexico

In 1980, in an era when decentralization was an important part of Mexico’s national development, a group of academics from UNAM founded, in the city of Guanajuato, an institution dedicated to basic research and high level education in the field of mathematics. Thus CIMAT (Centro de Investigación en Matemáticas A.C., Mathematics Research Center) was born.

Over the years CIMAT’s academic activity has been diversified and expanded to the point where it is now the leading research institution in its field outside of Mexico City, thanks to the high academic standards of its researchers, its vast scientific output and its ever increasing share of the international scientific scene. CIMAT organizes conferences, symposia, seminars and workshops, and offers bachelor’s, master's and doctorate degrees that are recognized internationally for their excellence.

CIMAT also has built strong links with the public, social, and business sectors of Mexican society, focusing on projects that help solve problems, particularly those related to technological innovation. In addition to its main campus in Guanajuato, CIMAT has established units in the cities of Aguascalientes, Zacatecas and Monterrey, where it contributes to the competitiveness and the growth of businesses and organizations, to the strengthening of mathematics skills in society at large, and to meeting the demand for a high level professional and scientific workforce.

04 11 2014

Deep Learning Tutorial by LISA Lab, University of Montreal

The tutorials presented here will introduce you to some of the most important deep learning algorithms and will also show you how to run them using Theano. Theano is a python library that makes writing deep learning models easy, and gives the option of training them on a GPU.

The algorithm tutorials have some prerequisites. You should know some python, and be familiar with numpy. Since this tutorial is about using Theano, you should read over the Theano basic tutorial first. Once you’ve done that, read through our Getting Started chapter, it introduces the notation, and [downloadable] datasets used in the algorithm tutorials, and the way we do optimization by stochastic gradient descent.

The purely supervised learning algorithms are meant to be read in order:

1. Logistic Regression, using Theano for something simple

2. Multilayer perceptron, introduction to layers

3. Deep Convolutional Network, a simplified version of LeNet5

The unsupervised and semi supervised learning algorithms can be read in any order (the auto encoders can be read independently of the RBM/DBN thread):

• Auto Encoders, Denoising Autoencoders, description of autoencoders

• Stacked Denoising Auto Encoders, easy steps into unsupervised pre training for deep nets

• Restricted Boltzmann Machines, single layer generative RBM model

• DeepBeliefNetworks unsupervisedgenerativepre trainingofstackedRBMsfollowedbysupervised fine tuning

Building towards including the mcRBM model, we have a new tutorial on sampling from energy models:

• HMC Sampling, hybrid (aka Hamiltonian) Monte Carlo sampling with scan()

Building towards including the Contractive auto encoders tutorial, we have the code for now:

• Contractive auto encoders code, There is some basic doc in the code.

Energy based recurrent neural network (RNN RBM):

• Modeling and generating sequences of polyphonic music

deeplearning.net

EMTECH Videos, MIT

A Place of Inspiration

It’s an opportunity to glimpse the future and begin to understand the technologies that matter and how they’ll change the face of business and drive the new global economy. It’s where technology, business, and culture converge. It’s the showcase for emerging technologies with the greatest potential to change our lives. It’s an access point to the most innovative people and companies in the world.

www.technologyreview.com

29 10 2014

edX: Introduction to Computational Thinking and Data Science

6.00.2x is aimed at students with some prior programming experience in Python and a rudimentary knowledge of computational complexity. We have chosen to focus on breadth rather than depth. The goal is to provide students with a brief introduction to many topics, so that they will have an idea of what’s possible when the time comes later in their career to think about how to use computation to accomplish some goal. That said, it is not a “computation appreciation” course. Students will spend a considerable amount of time writing programs to implement the concepts covered in the course. Topics covered include plotting, stochastic programs, probability and statistics, random walks, Monte Carlo simulations, modeling data, optimization problems, and clustering.

Tube19880 Videos

This channel contains amateur video of some of the many interesting technical presentations occurring in the San Francisco Bay Area, and mainly about software.

Silicon Valley Machine Learning Meetup (SVML) Videos, 2014

Tapping into biological networks to solve machine learning problems in cancer

Fingerprints from Words, Semantic Folding

Topological Data Analysis with Ayasdi

Learning to Rank search results

Python Tutorials by Jessica MacKellar (scroll down to access talks)

I am a startup founder, software engineer, and open source developer living in San Francisco, California.

I enjoy the Internet, networking, low level systems engineering, relational databases, tinkering on electronics projects, and contributing to and helping other people contribute to open source software.

"Be the change you wish to see in the world" may be clichéd, but what can I say, I believe in it. I am committed to applying my skills, in individual and collective efforts, to improve the world. Right now, this means I spend a lot of time volunteering, engaging technologists about education, and empowering effective people and initiatives in my capacity as a Director for the Python Software Foundation.

web.mit.edu

25 10 2014

20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), New York 2014

Videos

KDD 2014, a premier interdisciplinary conference, brings together researchers and practitioners from data science, data mining, knowledge discovery, large scale data analytics, and big data.

videolectures.net

Slides

www.kdd.org

Links before 24 oct 2014

Coursera

Machine Learning Stanford Course

This course provides a broad introduction to machine learning, datamining, and statistical pattern recognition. Topics include: (i) Supervised learning (parametric/non parametric algorithms, support vector machines, kernels, neural networks). (ii) Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning). (iii) Best practices in machine learning (bias/variance theory; innovation process in machine learning and AI). The course will also draw from numerous case studies and applications, so that you'll also learn how to apply learning algorithms to building smart robots (perception, control), text understanding (web search, anti spam), computer vision, medical informatics, audio, database mining, and other areas.

Pratical Machine Learning

One of the most common tasks performed by data scientists and data analysts are prediction and machine learning. This course will cover the basic components of building and applying prediction functions with an emphasis on practical applications. The course will provide basic grounding in concepts such as training and tests sets, overfitting, and error rates. The course will also introduce a range of model based and algorithmic machine learning methods including regression, classification trees, Naive Bayes, and random forests. The course will cover the complete process of building prediction functions including data collection, feature creation, algorithms, and evaluation.

Machine Learning Washington Course

Machine learning algorithms can figure out how to perform important tasks by generalizing from examples. This is often feasible and cost effective when manual programming is not. Machine learning (also known as data mining, pattern recognition and predictive analytics) is used widely in business, industry, science and government, and there is a great shortage of experts in it. If you pick up a machine learning textbook you may find it forbiddingly mathematical, but in this class you will learn that the key ideas and algorithms are in fact quite intuitive. And powerful!

Most of the class will be devoted to supervised learning (in other words, learning in which a teacher provides the learner with the correct answers at training time). This is the most mature and widely used type of machine learning. We will cover the main supervised learning techniques, including decision trees, rules, instances, Bayesian techniques, neural networks, model ensembles, and support vector machines. We will also touch on learning theory with an emphasis on its practical uses. Finally, we will cover the two main classes of unsupervised learning methods: clustering and dimensionality reduction. Throughout the class there will be an emphasis not just on individual algorithms but on ideas that cut across them and tips for making them work.

Core Concepts in Data Analysis (Higher School of Economics)

Learn both theory and application for basic methods that have been invented either for developing new concepts, principal components or clusters, or for finding interesting correlations, regression and classification. This is preceded by a thorough analysis of 1D and 2D data

This is an unconventional course in modern Data Analysis, Machine Learning and Data Mining. Its contents are heavily influenced by the idea that data analysis should help in enhancing and augmenting knowledge of the domain as represented by the concepts and statements of relation between them. According to this view, two main pathways for data analysis are summarization, for developing and augmenting concepts, and correlation, for enhancing and establishing relations. The term summarization embraces here both simple summaries like totals and means and more complex summaries: the principal components of a set of features and cluster structures in a set of entities. Similarly, correlation covers both bivariate and multivariate relations between input and target features including Bayes classifiers.

Neural Networks for Machine Learning

Neural Networks use learning algorithms that are inspired by our understanding of how the brain learns, but they are evaluated by how well they work for practical applications such as speech recognition, object recognition, image retrieval and the ability to recommend products that a user will like. As computers become more powerful, Neural Networks are gradually taking over from simpler Machine Learning methods. They are already at the heart of a new generation of speech recognition devices and they are beginning to outperform earlier systems for recognizing objects in images. The course will explain the new learning procedures that are responsible for these advances, including effective new proceduresr for learning multiple layers of nonlinear features, and give you the skills and understanding required to apply these procedures in many other domains.

Natural Language Processing

Natural language processing (NLP) deals with the application of computational models to text or speech data. Application areas within NLP include automatic (machine) translation between languages; dialogue systems, which allow a human to interact with a machine using natural language; and information extraction, where the goal is to transform unstructured text into structured (database) representations that can be searched and browsed in flexible ways. NLP technologies are having a dramatic impact on the way people interact with computers, on the way people interact with each other through the use of language, and on the way people access the vast amount of linguistic data now in electronic form. From a scientific viewpoint, NLP involves fundamental questions of how to structure formal models (for example statistical models) of natural language phenomena, and of how to design algorithms that implement these models.

Probabilistic Graphical Models

Uncertainty is unavoidable in real world applications: we can almost never predict with certainty what will happen in the future, and even in the present and the past, many important aspects of the world are not observed with certainty. Probability theory gives us the basic foundation to model our beliefs about the different possible states of the world, and to update these beliefs as new evidence is obtained. These beliefs can be combined with individual preferences to help guide our actions, and even in selecting which observations to make. While probability theory has existed since the 17th century, our ability to use it effectively on large problems involving many inter related variables is fairly recent, and is due largely to the development of a framework known as Probabilistic Graphical Models (PGMs). This framework, which spans methods such as Bayesian networks and Markov random fields, uses ideas from discrete data structures in computer science to efficiently encode and manipulate probability distributions over high dimensional spaces, often involving hundreds or even many thousands of variables. These methods have been used in an enormous range of application domains, which include: web search, medical and fault diagnosis, image understanding, reconstruction of biological networks, speech recognition, natural language processing, decoding of messages sent over a noisy communication channel, robot navigation, and many more. The PGM framework provides an essential tool for anyone who wants to learn how to reason coherently from limited and noisy observations.

Stanford Engineering Everywhere

SEE programming includes one of Stanford’s most popular engineering sequences: the three course Introduction to Computer Science taken by the majority of Stanford undergraduates, and seven more advanced courses in artificial intelligence and electrical engineering.

Introduction to Computer Science

Programming MethodologyCS106A

Programming AbstractionsCS106B

Programming ParadigmsCS107

Artificial Intelligence

Introduction to RoboticsCS223A

Natural Language ProcessingCS224N

Machine LearningCS229

Linear Systems and Optimization

The Fourier Transform and its ApplicationsEE261

Introduction to Linear Dynamical SystemsEE263

Convex Optimization IEE364A

Convex Optimization IIEE364B

Additional School of Engineering Courses

Programming Massively Parallel ProcessorsCS193G

iPhone Application ProgrammingCS193P

Seminars and Webinars

see.stanford.edu

EdX

Learning from data (Caltech)

This is an introductory course in machine learning (ML) that covers the basic theory, algorithms, and applications. ML is a key technology in Big Data, and in many financial, medical, commercial, and scientific applications. It enables computational systems to automatically learn how to perform a desired task based on information extracted from the data. ML has become one of the hottest fields of study today, taken up by undergraduate and graduate students from 15 different majors at Caltech. This course balances theory and practice, and covers the mathematical as well as the heuristic aspects.

Articifial Intelligence (BerkeleyX)

CS188.1x is a new online adaptation of the first half of UC Berkeley's CS188: Introduction to Artificial Intelligence. The on campus version of this upper division computer science course draws about 600 Berkeley students each year.

Artificial intelligence is already all around you, from web search to video games. AI methods plan your driving directions, filter your spam, and focus your cameras on faces. AI lets you guide your phone with your voice and read foreign newspapers in English. Beyond today's applications, AI is at the core of many new technologies that will shape our future. From self driving cars to household robots, advancements in AI help transform science fiction into real systems.

CS188.1x focuses on Behavior from Computation. It will introduce the basic ideas and techniques underlying the design of intelligent computer systems. A specific emphasis will be on the statistical and decision–theoretic modeling paradigm. By the end of this course, you will have built autonomous agents that efficiently make decisions in stochastic and in adversarial settings. CS188.2x (to follow CS188.1x, precise date to be determined) will cover Reasoning and Learning. With this additional machinery your agents will be able to draw inferences in uncertain environments and optimize actions for arbitrary reward structures. Your machine learning algorithms will classify handwritten digits and photographs. The techniques you learn in CS188x apply to a wide variety of artificial intelligence problems and will serve as the foundation for further study in any application area you choose to pursue.

Big Data and Social Physics (Ethics)

Social physics is a big data science that models how networks of people behave and uses these network models to create actionable intelligence. It is a quantitative science that can accurately predict patterns of human behavior and guide how to influence those patterns to (for instance) increase decision making accuracy or productivity within an organization. Included in this course is a survey of methods for increasing communication quality within an organization, approaches to providing greater protection for personal privacy, and general strategies for increasing resistance to cyber attack.

MIT OpenCourseWare (OCW)

OCW makes the materials used in the teaching of MIT's subjects available on the Web.

ocw.mit.edu

IPAM, Institute for Pure and Applied Mathematics, Videos, UCLA

IPAM records many of its lectures and makes them available to the public so that a wider audience may benefit from the scientific programs we offer. Since July 2012, IPAM has begun to record most of its lectures. You can access the lectures for a particular program or workshop (such as Materials Defects Tutorials) by following the program link listed below to the relevant workshop schedule. Each speaker is listed along with available slide shows and videos. For public lectures, the link will take you directly to the video. The programs and public lectures are listed in reverse chronological order.

Older videos play on Real Player only; recent videos will play on Flash supported browsers and software.

www.ipam.ucla.edu

Carnegie Mellon University Video resources

"The videos below are intended to serve as resources for our current students, and not as online learning materials for students outside of our program.", The Machine Learning Department

www.ml.cmu.edu

Metacademy Concept list and roadmap list

Metacademy is a community driven, open source platform for experts to collaboratively construct a web of knowledge. Right now, Metacademy focuses on machine learning and probabilistic AI, because that's what the current contributors are experts in. But eventually, Metacademy will cover a much wider breadth of knowledge, e.g. mathematics, engineering, music, medicine, computer science…

www.metacademy.org

Harvard, Advanced Machine Learning, Fall 2013 (Free access to most of videos)

This course is about learning to extract statistical structure from data, for making decisions and predictions, as well as for visualization. The course will cover many of the most important math ematical and computational tools for probabilistic modeling, as well as examine specific models from the literature and examine how they can be used for particular types of data. There will be a heavy emphasis on implementation. You may use Matlab, Python or R. Each of the five assign ments will involve some amount of coding, and the final project will almost certainly require the running of computer experiments.

www.seas.harvard.edu

Oxford University, Nando de Freitas video lectures

I am a machine learning professor at UBC. I am making my lectures available to the world with the hope that this will give more folks out there the opportunity to learn some of the wonderful things I have been fortunate to learn myself. Enjoy.

Caltech University, Learning from Data

Free, introductory Machine Learning online course (MOOC)

Taught by Caltech Professor Yaser Abu Mostafa [article]

Lectures recorded from a live broadcast, including Q&A

Prerequisites: Basic probability, matrices, and calculus

8 homework sets and a final exam

Discussion forum for participants

Topic by topic video library for easy review

work.caltech.edu

University College London Discovery

UCL Discovery showcases UCL's research publications, giving access to journal articles, book chapters, conference proceedings, digital web resources, theses and much more, from all UCL disciplines. Where copyright permissions allow, a full copy of each research publication is directly available from UCL Discovery.

You can search or browse UCL Discovery, see the most downloaded publications, and keep up to date with the latest UCL research by RSS or even on Twitter.

discovery.ucl.ac.uk

University College London, Supervised Learning Course

The course covers supervised approaches to machine learning. It starts by probabilistic pattern recognition followed by an in depth introduction to various supervised learning algorithms such as Least Squares, Lasso, Perceptron Algorithm, Support Vector Machines and Boosting.

www0.cs.ucl.ac.uk

Yann LeCun’s Publications

My main research interests are Machine Learning, Computer Vision, Mobile Robotics, and Computational Neuroscience. I am also interested in Data Compression, Digital Libraries, the Physics of Computation, and all the applications of machine learning (Vision, Speech, Language, Document understanding, Data Mining, Bioinformatics).

yann.lecun.com

Francis Bach, Ecole Normale Superieure, Courses and Exercises with solutions (English French)

Spring 2014: Statistical machine learning, Master M2 "Probabilites et Statistiques", Universite Paris Sud (Orsay)

Fall 2013: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

Spring 2013: Statistical machine learning, Master M2 "Probabilites et Statistiques", Universite Paris Sud (Orsay)

Spring 2013: Statistical machine learning, Filiere Math/Info, L3, Ecole Normale Superieure (Paris)

Fall 2012: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

Spring 2012: Statistical machine learning, Filiere Math/Info, L3, Ecole Normale Superieure (Paris)

Spring 2012: Statistical machine learning, Master M2 "Probabilites et Statistiques", Universite Paris Sud (Orsay)

Fall 2011: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

Spring 2011: Statistical machine learning, Master M2 "Probabilites et Statistiques", Universite Paris Sud (Orsay)

Fall 2010: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

Spring 2010: Statistical machine learning, Master M2 "Probabilites et Statistiques", Universite Paris Sud (Orsay)

Fall 2009: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

Fall 2008: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

May 2008: Probabilistic modelling and graphical models: Enseignement Specialise, Ecole des Mines de Paris

Fall 2007: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

May 2007: Probabilistic modelling and graphical models: Enseignement Specialise, Ecole des Mines de Paris

Fall 2006: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

Fall 2005: An introduction to graphical models, Master M2 "Mathematiques, Vision, Apprentissage", Ecole Normale Superieure de Cachan

www.di.ens.fr

Videolectures.net

VideoLectures.NET is an award winning free and open access educational video lectures repository. The lectures are given by distinguished scholars and scientists at the most important and prominent events like conferences, summer schools, workshops and science promotional events from many fields of Science. The portal is aimed at promoting science, exchanging ideas and fostering knowledge sharing by providing high quality didactic contents not only to the scientific community but also to the general public. All lectures, accompanying documents, information and links are systematically selected and classified through the editorial process taking into account also users' comments.

videolectures.net

MLSS Machine Learning Summer Schools Videos

MLSS Videos from 2004 to 2012

videolectures.net

MLSS Videos 2012

MLSS Videos 2012

Max Planck Institute for Intelligent Systems Tubingen, MLSS Videos 2013

Our goal is to understand the principles of Perception, Action and Learning in autonomous systems that successfully interact with complex environments and to use this understanding to design future systems. The Institute studies these principles in biological, computational, hybrid, and material systems ranging from nano to macro scales.We take a highly interdisciplinary approach that combines mathematics, computation, material science, and biology.

The MPI for Intelligent Systems has campuses in Stuttgart and Tübingen. Our Stuttgart campus has world leading expertise in small scale intelligent systems that leverage novel material science and biology. The Tübingen campus focuses on how intelligent systems process information to perceive, act and learn.

MLSS 2014 Pittsburgh + Alex Smola's playlist

GoogleTechTalks

Machine Learning

Deep Learning

Udacity Opencourseware

Supervised Learning (select "View Courseware" for free access)

Why Take This Course?

In this course, you will gain an understanding of a variety of topics and methods in Supervised Learning. Like function approximation in general, Supervised Learning prompts you to make generalizations based on fundamental assumptions about the world.

Michael: So why wouldn't you call it "function induction?"

Charles: Because someone said "supervised learning" first.

Topics covered in this course include: Decision trees, neural networks, instance based learning, ensemble learning, computational learning theory, Bayesian learning, and many other fascinating machine learning concepts.

www.udacity.com

Unsupervised Learning (select "View Courseware" for free access)

Why Take This Course?

You will learn about and practice a variety of Unsupervised Learning approaches, including: randomized optimization, clustering, feature selection and transformation, and information theory.

You will learn important Machine Learning methods, techniques and best practices, and will gain experience implementing them in this course through a hands on final project in which you will be designing a movie recommendation system (just like Netflix!).

www.udacity.com

Reinforcement Learning (select "View Courseware" for free access)

Why Take This Course?

You will learn about Reinforcement Learning, the field of Machine Learning concerned with the actions that software agents ought to take in a particular environment in order to maximize rewards.

Michael: Reinforcement Learning is a very popular field.

Charles: Perhaps because you're in it, Michael.

Michael: I don't think that's it.

In this course, you will gain an understanding of topics and methods in Reinforcement Learning, including Markov Decision Processes and Game Theory. You will gain experience implementing Reinforcement Learning techniques in a final project.

In the final project, we’ll bring back the 80's and design a Pacman agent capable of eating all the food without getting eaten by monsters.

www.udacity.com

Mathematicalmonk Machine Learning

Videos about math, at the graduate level or upper level undergraduate.

Judea Pearl Symposium

Judea Pearl (born 1936) is an Israeli born American computer scientist and philosopher, best known for championing the probabilistic approach to artificial intelligence and the development of Bayesian networks (see the article on belief propagation). He is also credited for developing a theory of causal and counterfactual inference based on structural models (see article on causality). He is the 2011 winner of the ACM Turing Award, the highest distinction in computer science, "for fundamental contributions to artificial intelligence through the development of a calculus for probabilistic and causal reasoning". (source Wikipedia)

Machine Learning Reading Group, Indian Institute of Science

Focus Areas: Machine Learning & Convex Optimization

clweb.csa.iisc.ernet.in

SIGDATA, Indian Institute of Technology Kanpur

www.cse.iitk.ac.in

Hakka Labs

Hakka Labs is passionate about helping professional software engineers level up in their careers. Our content, events & community have grown by leaps and bounds since our humble origin when we launched as a Tumblr blog in 2011.

We believe that "software is eating the world" and our passion is in building valuable resources and community for startup oriented software engineers, the folks that will power innovation and disrupt industries, and ultimately shape our future.

Hakka originally launched in SF Bay & NYC and rapidly built relationships with the top companies, CTOs and tech influencers in these key areas. We have deep connections to the software engineering worlds on both coasts and often invite groups of CTOs and engineers to our office in Soho, or meet with them at engineering events that we either run or participate in.

We're also currently up & running in Berlin & Moscow, and plan to continue to rapidly expand worldwide. Not too shabby for a scrappy startup with a small marketing budget!

www.hakkalabs.co

Open Yale Course

Game Theory

Each course includes a full set of class lectures produced in high quality video accompanied by such other course materials as syllabi, suggested readings, exams, and problem sets. The lectures are available as downloadable videos, and an audio only version is also offered. In addition, searchable transcripts of each lecture are provided.

oyc.yale.edu

Deep Learning

Deep Learning is a new area of Machine Learning research, which has been introduced with the objective of moving Machine Learning closer to one of its original goals: Artificial Intelligence.

This website is intended to host a variety of resources and pointers to information about Deep Learning. In these pages you will find

• a reading list,

• links to software,

• datasets,

• a list of deep learning research groups and labs,

• a list of announcements for deep learning related jobs (job listings),

• as well as tutorials and cool demos.

For the latest additions, including papers and software announcement, be sure to visit the Blog section and subscribe to our RSS feed of the website. Contact us if you have any comments or suggestions!

www.deeplearning.net

deeplearning.net

BigDataWeek Videos

Big Data Week is one of the most unique global platforms of interconnected community events focusing on the social, political, technological and commercial impacts of Big Data. It brings together a global community of data scientists, data technologies, data visualisers and data businesses spanning six major commercial, financial, social and technological sectors.

Neural Information Processing Systems Foundation (NIPS) Video resources

The Foundation: The Neural Information Processing Systems (NIPS) Foundation is a non profit corporation whose purpose is to foster the exchange of research on neural information processing systems in their biological, technological, mathematical, and theoretical aspects. Neural information processing is a field which benefits from a combined view of biological, physical, mathematical, and computational sciences.

The primary focus of the NIPS Foundation is the presentation of a continuing series of professional meetings known as the Neural Information Processing Systems Conference, held over the years at various locations in the United States, Canada and Spain.

Hong Kong Open Source Conference 2013 (English&Chinese)

Wang Leung Wong

The Vice Chairperson of the Hong Kong Linux User Group

This channel will post the videos of my life and opensource events in Hong Kong.

Hong Kong Linux User Group: http://linux.org.hk

Facebook: https://www.facebook.com/groups/hklug/

ICLR 2014 Videos

It is well understood that the performance of machine learning methods is heavily dependent on the choice of data representation (or features) on which they are applied. The rapidly developing field of representation learning is concerned with questions surrounding how we can best learn meaningful and useful representations of data. We take a broad view of the field, and include in it topics such as deep learning and feature learning, metric learning, kernel learning, compositional models, nonlinear structured prediction, and issues regarding non convex optimization.

Despite the importance of representation learning to machine learning and to application areas such as vision, speech, audio and NLP, there is currently no common venue for researchers who share a common interest in this topic. The goal of ICLR is to help fill this void.

ICLR 2014 will be a 3day event from April 14th to April 16th 2014, in Banff, Canada. The conference will follow the recently introduced open reviewing and open publishing publication process, which is explained in further detail here: Publication Model.

ICLR 2013 Videos

ICLR 2013 will be a 3day event from May 2nd to May 4th 2013, co located with AISTATS2013 in Scottsdale, Arizona. The conference will adopt a novel publication process, which is explained in further detail here: Publication Model.

sites.google.com

Machine Learning Conference Videos

Events matching your search:

• ICML 2011

• Sixth Annual Machine Learning Symposium

• 1st Lisbon Machine Learning School

• Copulas in Machine Learning Workshop 2011

• NIPS 2011 Workshop on Integrating Language and Vision

• Machine Learning in Computational Biology (MLCB) 2011

• Learning Semantics Workshop

• Sparse Representation and Low rank Approximation

• The 4th International Workshop on Music and Machine Learning: Learning from Musical Structure

• Big Learning: Algorithms, Systems, and Tools for Learning at Scale

• ICML 2012 Oral Talks (International Conference on Machine Learning)

• Big Data Meets Computer Vision: First International Workshop on Large Scale Visual Recognition and Retrieval

• 2nd Workshop on Semantic Perception, Mapping and Exploration (SPME)

• ICML 2012 Workshop on Representation Learning

• Inferning 2012: ICML Workshop on interaction between Inference and Learning

• Object, functional and structured data: towards next generation kernel based methods, ICML 2012 Workshop

• Tutorial on Statistical Learning Theory in Reinforcement Learning and Approximate Dynamic Programming

• Tutorial on Causal inference, conditional independences and beyond

• ICML 2012 Tutorial on Prediction, Belief, and Markets

• PAC Bayesian Analysis in Supervised, Unsupervised, and Reinforcement Learning

• Performance Evaluation for Learning Algorithms: Techniques, Application and Issues

• 2nd Lisbon Machine Learning School (2012)

• OpenCV using Python

• Big Learning : Algorithms, Systems, and Tools

• NIPS 2012 Workshop on Log Linear Models

• Machine Learning in Computational Biology (MLCB) 2012

• NYU Course on Big Data, Large Scale Machine Learning

• Sixteenth International Conference on Artificial Intelligence and Statistics (AISTATS) 2013

• International Conference on Learning Representations (ICLR) 2013

• ICML 2013 Plenary Webcast

• NYU Course on Deep Learning (Spring 2014)

techtalks.tv

• NYU Course on Machine Learning and Computational Statistics 2014

techtalks.tv

MOOC/OPENCOURSEWARE FRENCH

Hugo Larochelle, Apprentissage automatique, French Canadian

1,592 subscribers

128,894 views (source 07 06 2014)

College de France, Mathematics and Digital Science, French

One of the Collège de France's missions is to promote French research and thought abroad, and to participate in intel lectual debates on major world issues. The institution therefore participates in international exchange through its teaching and the dissemination of knowledge, as well as through the research programmes involving its Chairs and laboratories. The fact that one fifth of the professors are currently from abroad, confirms the Collège de France's wid ening research and education policy.

This policy of international openness translates into:

• Collège de France professors' teaching missions abroad

• Lectures and lecture series by visiting professors

• Junior Visiting Researchers scheme

• Lecture series and symposia abroad

• Internet broadcasts

www.college de france.fr

MOOC/OPENCOURSEWARE CHINESE

Yeeyan Coursera Chinese Classroom

Google Translation from Chinese (Simplified Han) to English

Welcome to Yeeyan × Coursera Chinese classroom.

In this always have a small partner to accompany the classroom, you can:

join collaborative translation;

exchange ideas;

enrollment became class representative;

punch seek supervision;

......

Finally, welcome to drying out your certificate, either × Coursera joint Yeeyan Translator's Certificate or Certificate of Coursera course, you are overcome my own life winner!

coursera.yeeyan.org

Hong Kong Open Source Conference 2013

Wang Leung Wong

The Vice Chairperson of the Hong Kong Linux User Group

This channel will post the videos of my life and opensource events in Hong Kong.

Hong Kong Linux User Group: http://linux.org.hk

Facebook: https://www.facebook.com/groups/hklug/

MOOC/OPENCOURSEWARE HEBREW

Open University of Israel

האוניברסיטה הפתוחה היא ייחודית בנוף האקדמי בישראל.

היא דומה לאוניברסיטאות האחרות בחתירתה למצוינות ובשקידתה על איכות למדנית ומדעית גבוהה, אך היא שונה מהן במבנה הארגוני שלה, בשיטות ההוראה שלה, במערך תכניות הלימודים ובדרישותיה מן המועמדים הפונים להירשם לקורסים שלה.

האוניברסיטה הפתוחה, כשמה כן היא. היא פותחת את שעריה, בלא תנאים מוקדמים ובלי דרישות קדם, הן בפני מי שמבקשים ללמוד קורסים בודדים או חטיבות קורסים, הן בפני מי שמעוניינים ללמוד תכנית לימודים מלאה לתואר "בוגר אוניברסיטה".