The Machine Learning Reading Group (MLRG) meets regularly (usually weekly) to discuss research topics on a particular sub-field of Machine Learning.

You can receive announcements about the reading group by joining our mailing list. To join the mailing list, please use an academic email address and send an email to majordomo@cs.ubc.ca with an empty subject line and with the following message body: "subscribe mlrg-l YOUR-EMAIL-ADDRESS". If you use a non-academic email address, we would have to verify it which could delay your subscription process.

Winter term 1 2018 - Reinforcement Learning 2Every Monday in room ICICS 146 at 5:00 PM | ||

Date | Presenter | Topic |

Oct 15 | Mark | Motivation/Overview - [pdf slides] |

Oct 22 | Yifan | Bayesian RL - [pdf slides] |

Oct 29 | Christian | Useful Uncertainties in Reinforcement Learning - [pdf slides] |

Nov 5 | Sharan | Introduction to Bandits - [pdf slides] |

Nov 12 | Cancelled | |

Nov 19 | Aaron | |

Nov 26 | Boyan | |

Dec 3 | Wilder | |

Dec 10 | Mehrdad | |

Dec 17 | Vadan |

Summer 2018 - Every Tuesday in room ICICS 146 at 3:00 PM | ||

Date | Presenter | Topic |

May 08 | Emtiyaz Khan | Fast and Scalable Bayesian Deep Learning by Weight-Perturbation in Adam |

May 15 | Geoff Roeder | Better Inference through Lower-Variance Stochastic Gradients |

May 22 | Brendan Juba | Learning Abduction Under Partial Observability |

Winter term 2 2018 - Parallel and Distributed Machine LearningEvery Tuesday in room ICICS 146 at 5:00 PM | ||

Date | Presenter | Topic |

Jan 30 | Mark Schmidt | Motivation - [pdf slides] |

Feb 6 | Yasha | Distributed file systems |

Feb 13 | Michael | Asynchronous stochastic gradient |

Feb 27 | Sharan | Synchronous stochastic gradient - [pdf slides] |

Mar 6 | Julie | Parallel coordinate optimization - [pdf slides] |

Mar 13 | Devon | Decentralized gradient |

Mar 20 | Wu | Decomposition methods |

Mar 27 | Reza | Asynchronous/distributed SAG/SDCA/SVRG |

Apr 3 | Vaden | Randomized Newton and least squares on the cloud |

Apr 10 | Nasim | Parallel tempering and distributed particle filtering |

Apr 17 | Alireza | Distributed deep networks |

Apr 24 | Raunak | Blockchain-based distributed learning |

Winter term 1 2017 - Deep Learning Meets Graphical ModelsEvery Tuesday in room ICICS 146 at 4:00 PM | ||

Date | Presenter | Topic |

Sep 26 | Mark | Motivation/overview - [pdf slides] |

Oct 3 | Issam | FCNs and CRFs |

Oct 10 | Julieta | RNNs |

Oct 17 | Michael | Bayesian neural nets 1: sampling |

Oct 24 | Jason | Bayesian neural nets 2: variational |

Oct 31 | Devon | Variational autoencoders 1: basics/ - [pdf slides] |

Nov 7 | Sharan | Variational autoencoders 2: variations - [pdf slides] |

Nov 14 | Mohamed | Generative adversarial networks 1: basics |

Nov 21 | Alireza | Generative adversarial networks 2: variations |

Nov 28 | Raunak | Beyond generative adversarial networks/ - [pdf slides] |

Summer 2017 - Online, Active, and Causal learningEvery Tuesday in room ICICS 146 at 4:00 PM, | ||

Date | Presenter | Topic,,, |

Jun 6 | Mark Schmidt | Motivation/overview, perceptron, follow the leader. - [pdf slides] |

Jun 13 | Julie | Online convex optimization, mirror descent - [pdf slides] |

Jun 20 | Alireza | Multi-armed bandits, contextual bandits - [pdf slides] |

Jun 27 | Michael | Heavy hitters,,, |

Jul 4 | Raunak | Regularized FTL, AdaGrad, Adam, online-to-batch - [pdf slides] |

Jul 11 | Glen | Best-arm identification, dueling bandits,, |

Jul 18 | Nasim | Uncertainty sampling, variance/error reduction, QBC - [pdf slides] |

Jul 25 | Mohamed | Planning, A/B testing, Optimal experimental design, |

Aug 1 | Sanna | Randomized controlled trials, do-calculus - [pdf slides] |

Aug 8 | Issam | Granger causality, independent component analysis,, |

Aug 22 | Eric | Counterfactuals - [pdf slides] |

Aug 29 | Jason | Instrumental variables,,, |

Winter term 2 2017 - Reinforcement LearningEvery Tuesday in room ICICS 146 at 5:00 PM,, | ||

Date | Presenter | Topic,, |

Jan 10 | Mark Schmidt | Motivation/Overview - [pdf slides] |

Jan 17 | Nasim | MDPs (policy iteration, value iteration), |

Jan 24 | Julie | Monte Carlo (estimators, on-policy/off-policy learning) - [pdf slides] |

Jan 31 | Raunak | Temporal Difference Learning,, |

Feb 7 | Jennifer | Multi-Step Bootstrapping/ - [pdf slides] |

Feb 14 | Michael | Function Approximation, TD-Gammon, |

Feb 21 | Cancelled,, | |

Feb 28 | Ricky | Planning, Control with Approximation, and Eligibility Traces |

Mar 7 | Issam | Optimal control, flying helicopters, |

Mar 14 | Sharan | POMDPs - [pdf slides] |

Mar 21 | Jason | Policy gradients, Monte Carlo tree search, and AlphaGo |

Mar 28 | Julieta | Value-Iteration Networks,, |

Apr 4 | Glen | RL in Practice,, |

Apr 11 | Michiel | Perspectives on Reinforcement Learning for Locomotion Skills,, |

Apr 25 | Issam | Connection between Generative Adversarial Networks and Inverse Reinforcement Learning,, |

Winter term 1 2016 - Deep LearningEvery Wednesday in room ICICS 146 at 5:00 PM | ||

Date | Presenter | Topic |

Sep 21 | Mark Schmidt | Introduction - [pdf slides] |

Sep 28 | Julie | Feedforward neural nets, backpropagation - [pdf slides] |

Oct 5 | Mohamed | Network-independent tricks - [pdf slides] |

Oct 12 | Issam | ImageNet tricks |

Oct 19 | Jason | Graphical models - [pdf slides] |

Oct 26 | Saif | Artistic style transfer - [pdf slides] |

Nov 2 | Nasim | Recurrent neural nets - [pdf slides] |

Nov 9 | Stephen/Kevin | Recurrent neural nets 2 |

Nov 16 | Ricky | Variational autoencoders and Bayesian dark knowledge |

Nov 23 | Reza | Generative adversarial networks |

Nov 30 | Alireza | Memory nets, neural Turing, stack-augmented RNNs |

Summer term 2016 - MiscellaneousEvery Wednesday in room ICCS146 at 5:00 PM | ||

Date | Presenter | Topic |

May 25 | Mark Schmidt | Introduction to Summer topics - [pdf slides] |

Jun 1 | No meeting | UAI camera-ready deadline |

Jun 8 | Sharan | Spectral Methods (1) - [pdf slides] |

Jun 15 | Geoff | Spectral Methods (2) - [pdf slides] |

Jun 22 | Chris | Relational Models |

Jun 29 | Saif | Submodularity - [pdf slides] |

Jul 6 | Nasim | Grammars - [pdf slides] |

Jul 13 | Eviatar | Continuous graphical models - [pdf slides] |

Jul 20 | Steven and Kevin | Gaussian Copulas - [pdf slides] |

Jul 27 | Issam | Large-scale kernels methods (1) |

Aug 3 | Julietta | Large-scale kernels methods (2) |

Aug 10 | Alireza | Changepoint detection (1) |

Aug 17 | Mohamed | Changepoint detection (2) |

Aug 24 | Julie | Independent component analysis (1) |

Aug 31 | Ricky | Independent component analysis (2) |

Winter term 2 2016 - Crash course on Bayesian methodsEvery Wednesday in room ICICS 146 at 5:00 PM | ||

Date | Presenter | Topic |

Jan 06 | Mark Schmidt | Introduction to Bayesian methods - [pdf slides] |

Jan 13 | Nasim | Conjugate Priors, Non-Informative Priors - [pdf slides] |

Jan 20 | Geoff | Hierarchical Modeling and Bayesian Model Selection - [pdf slides] |

Jan 27 | Issam | Gaussian Processes and Empirical Bayes - [pdf slides] |

Feb 3 | Ricky | Basic Monte Carlo Methods - [pdf slides] |

Feb 10 | Jason | MCMC - [website link] |

Feb 24 | Michael | Bayesian Optimization - [pdf slides] |

Mar 2 | Sharan | Variational Bayes - [pdf slides] |

Mar 9 | Reza | Stochastic Variational Inference - [pdf slides] |

Mar 16 | Mark | Non-Parametric Bayes 1 - [pdf slides] |

Mar 23 | Reza | Non-Parametric Bayes 2 |

Apr 6 | Julieta | Sequential Monte Carlo and Population MCMC |

Apr 13 | Rudy | Reversible-Jump MCMC |

Apr 20 | Alireza | Approximate Bayesian Computation - [pdf slides] |

Winter term 1 2015 - Crash course on optimizationEvery Tuesday in room X836 at 5:00 PM | ||

Date | Presenter | Topic |

Sep 22 | Mark Schmidt | Introduction to convex optimization - [pdf slides] |

Sep 29 | Mark Schmidt | First-Order Methods - [pdf slides] |

Oct 06 | Julieta | Stochastic Subgradient - [pdf slides] |

Oct 13 | Mohamed | Minimizing Finite Sums - [pdf slides] |

Oct 20 | Jason | Proximal-Gradient - [pdf slides] |

Oct 27 | Ives | Frank-Wolfe, ADMM - [pdf slides] |

Nov 03 | Julie | Coordinate Descent - [pdf slides] |

Nov 10 | Sharan | Online Convex Optimization - [pdf slides] |

Nov 17 | Mark Schmidt | Multi-Level Methods - [pdf slides] |

Nov 24 | Issam | Non-Convex Rates - [pdf slides] |

Dec 01 | Issam | Parallel/Distributed - [pdf slides] |

Dec 08 | (NIPS) | |

Dec 15 | Alireza | Deep Learning Local Optima - [pdf slides] |

Summer term 2 2015 - Crash course on graphical modelsRoom ICICS 238 at 11:00 AM | ||

Date | Presenter | Topic |

Aug 17 | Mark Schmidt | Why learn about graphical models? - [pdf slides] |

Aug 18 | Mark Schmidt | Inference in Chains and Trees - [pdf slides] |

Aug 19 | Julie | Conditional Inference and Cutset Conditioning - [pdf slides] |

Aug 20 | Mehran | Junction Tree - [pdf slides] |

Aug 21 | Alireza | Semi-Markov/Graph Cuts - [pdf slides] |

Aug 24 | Mark Schmidt | MRF/CRF - [pdf slides] |

Aug 25 | Julieta | ICM/Block/Alpha - [pdf slides] |

Aug 26 | Jason | MCMC/Herding - [pdf slides] |

Aug 27 | Ankur | Hidden/RBM/Younes - [pdf slides] |

Aug 28 | Sharan | Structure Learning - [pdf slides] |

Aug 31 | Mark Schmidt | Variational/MF - [pdf slides] |

Sep 1 | Nasim | Bethe/Kikuchi - [pdf slides] |

Sep 2 | Reza | TRBP/Convex - [pdf slides] |

Sep 3 | Issam | LP/SDP - [pdf slides] |

Sep 4 | Mark Schmidt | SSVM/BCFW - [pdf slides] |