Welcome to the Heriot-Watt University Machine Learning Reading Group!

Time: Tuesdays 2.15pm

Place: EM 1.58


 18.Oct.2016 Transfer LearningSimon Keizer  Proactive Transfer Learning for Heterogeneous Feature and Label Spaces - Shane Moon and Jaime Carbonell



 23.Feb.2016 Multi-domain statistical dialogue managementSimon Keizer
 Multi-agent learning in multi-domain spoken dialogue systems (NIPS'15 Workshop on SLU and Interaction)
 Policy Committee for adaptation in multi-domain spoken dialogue sytems (ASRU'15)
 15.Mar.2016 Long Short-Term Memory NetworksEli Sheppard  Understanding LSTM Networks
 LSTM: A Search Space Odyssey
 22.Mar.2016 Long Short-Term Memory NetworksXingxing Zhang (UEDIN)
 Top-down Tree Long Short-Term Memory Networks (arXiv, 2015)
 5.Apr.2016 Neural Natural Language GenerationSimon Keizer
 Semantically Conditioned LSTM-based Natural Language Generation for Spoken Dialogue Systems (EMNLP'15 best paper)
 Stochastic Language Generation in Dialogue using Recurrent Neural Networks with Convolutional Sentence Reranking (SIGdial'15 best paper)
  12.Apr.2016   Gaussian Processes for Reinforcement LearningFrank Broz   Reinforcement learning with Gaussian processes
 10.May.2016 Gaussian Processes for POMDP optimisation Simon Keizer  Gaussian Processes for POMDP-based Dialogue Manager optimisation
 17.May.2016 Deep Reinforcement Learning Ioannis Efstathiou  Language Understanding for Text-based Games Using Deep Reinforcement Learning
 24.May.2016 Neural Natural Language Generation Shawn Wen (UCAM)   Beyond Conditional LM: Neural Network Language Generation for Dialogue Systems (see also the 5 April session)


 26.Jan.2015 Deep Learning for Perception
Heriberto Cuayáhuitl
Deep Speech: Scaling up end-to-end speech recognition
Multimodal Deep Learning
 Restricted Boltzmann Machines
Ingo Keller
 Training Restricted Boltzmann Machines: An Introduction
 23.Feb.2015 Multiagent Systems
Stefano Albrecht*

On Convergence and Optimality of Best-Response Learning with Policy Types in Multiagent Systems
An Empirical Study on the Practical Impact of Prior Beliefs over Policy Types
 Weakly-Supervised Semantic Parsing
Siva Reddy*

Large-Scale Semantic Parsing without Question-Answer Pairs
 Restricted Boltzmann Machines (continuation)Ingo Keller Training Restricted Boltzmann Machines: An Introduction
 27.Apr.2015 Non-parametric Bayesian learning & Interactive POMDPsSimon Keizer Panella & Gmytrasiewicz - Nonparametric Bayesian Learning of Other Agents' Policies in Interactive POMDPs
 13.May.2015 Statistical Testing Dimitra Gkatzia P Value and the Theory of Hypothesis Testing: An Explanation for New Researchers
Using Effect Size—or Why the P Value Is Not Enough
 8.Jun.2015 Deep Reinforcement Learning Oliver Lemon slidesHuman-Level Control Through Deep Reinforcement Learning
Playing Atari with Deep Reinforcement Learning
 22.Jun.2015 Convolutional Neural Nets
 Pawel Swietojanski
 Convolutional neural networks for distant speech recognition
Neural networks for distant speech recognition
 July/August SUMMER BREAK   
 28.Sept.2015 Deep Reinforcement Learning Heriberto Cuayáhuitl  Strategic Dialogue Management via Deep Reinforcement Learning
 12.Oct.2015 Symbol Grounding Arash Eshghi slides Learning Grounded Relational Symbols from Continuous Data for Abstract Reasoning
 26.Oct.2015 P-values and Statistical Testing Simon Keizer  Toward Evidence-Based Medical Statistics. 2: The P Value Fallacy
Toward Evidence-Based Medical Statistics. 2: The Bayes Factor
 23.Nov.2015 Incremental Bayesian Learning of Semantic Categories and their Features Lea Frermann  Incremental Bayesian Learning of Semantic Categories and their Features
 14.Dec.2015 NIPS report Heriberto Cuayáhuitl  Deep Reinforcement Learning Workshop
Deep Learning Symposium
Main Conference Papers 


 24.Feb.2014 Deep neural networks for speech recognitionLiang Liu* Deep Neural Networks for Speech Recognition
 05.May.2014 Deep Learning for State Tracking in Spoken Dialogue SystemsCallum Main  (dry run for talk at Deep Learning Workshop)
 09.June.2014 Natural Actor Critic Simon Keizer slides Peters & Schaal - Natural Actor-Critic
 16.June.2014 POMDP & SARSOPZhuoran Wang slides SARSOP: Efficient Point-Based POMDP Planning by Approximating Optimally Reachable Belief Spaces
 23.June.2014 Particle FilteringHeriberto Cuayáhuitl slides Particle Filters in Robotics
 Using Particle Filters to Track Dialogue State
 30.June.2014 Gaussian Processes in Reinforcement LearningArash Eshghi slides Gaussian Processes for Machine Learning (Chapters 1 & 2)
 Gaussian processes for POMDP-based dialogue manager optimisation
 07.July.2014 Discussion Session
Heriberto Cuayáhuitl  Machine Learning that Matters
 29.Sep.2014 Inverse Language Understanding Heriberto Cuayáhuitl  Asking for Help Using Inverse Semantics
 20.Oct.2014 Learning Word Meanings Oliver Lemon slides Learning Perceptually Grounded Word Meanings from Unaligned Parallel Data
 27.Oct.2014 Learning Semantics Arash Eshghi slides Combined Distributional and Logical Semantics
Combining Formal and Distributional Models of Temporal and Intentional Semantics
 17.Nov.2014 Learning Grounded Meanings  Carina Silberer*  Learning Grounded Meaning Representations with Autoencoders
 01.Dec.2014 Probabilistic Topic Modelling Simon Keizer slides
 Probabilistic Topic Models
Latent Dirichlet Allocation
* Visitors from the University of Edinburgh.

 21.Jan.2013Security in Machine Learning SystemsPhilippe De Wildeslides
 The security of machine learning
 28.Jan.2013 HMMs 2 (Expectation-Maximization) Murat Uney A Gentle Tutorial of the EM Algorithm and its Application to Parameter Estimation for Gaussian Mixture and Hidden Markov Models
Maximum Likelihood from Incomplete Data via the EM Algorithm
A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition
 04.Feb.2013 Graphical Models 2 (inference+parameter learning) Simon Keizer

Bishop: Chapter 8, Section 4
Local Computations with Probabilities on Graphical Structures and Their Application to Expert Systems - Lauritzen & Spiegelhalter, 1988
A Tutorial on Learning with Bayesian Networks - Heckerman, 2008
 11.Feb.2013 Spectral Clustering Heriberto Cuayáhuitl slidesA Tutorial on Spectral Clustering
Survey of Clustering Algorithms
On Spectral Clustering: Analysis and an Algorithm
Normalized Cuts and Image Segmentation
 18.Feb.2013 Linear Models for Regression Srini Janarthanam  Andrew Ng's notes on Supervised Learning and Discriminative Algorithms
 25.Feb.2013 Transfer Learning 1 Ben Rosman* slidesTransfer learning for reinforcement learning domains: A survey
Probabilistic policy reuse in a reinforcement learning agent
What good are actions? Accelerating learning using learned action priors
 04.Mar.2013 Ensembles Tom Larkworthy Classification and regression by randomForest
 11.Mar.2013Reinforcement Learning 2 (Linear Function Approximation)Heriberto Cuayáhuitlslides (long)
slides (short)
 Algorithms for reinforcement learning
 18.Mar.2013Boltzmann Machines and their extensions
 Ali Eslami* slidesTraining Products of Experts by Minimizing Contrastive Divergence
Deep Boltzmann Machines
The Shape Boltzmann Machine
 25.Mar.2013 Learning from Demonstration (postponed)
 Heriberto Cuayáhuitl  
 01.Apr.2013 [Easter Monday]   
 08.Apr.2013 Modeling Using Integer Linear Programming Kristian Woodsend* slidesGlobal Inference Using Integer Linear Programming
ILP-Based Reasoning for Weighted Abduction
 15.Apr.2013 Dirichlet Process and Hierarchical Dirichlet Process
 Zhuoran Wang slidesDirichlet Process
Hierarchical Dirichlet Processes
 22.Apr.2013 Incremental Learning (postponed)
  Nick Taylor  
 29.Apr.2013 Optimal Control of Variable Stiffness Jun Nakanishi* slidesIterative linearization methods for approximately optimal control and estimation of non-linear stochastic systems
Exploiting Passive Dynamics with Variable Stiffness Actuation in Robot Brachiation
 06.May.2013 Principal Component Analysis (postponed)
 Amol Deshmukh  
 13.May.2013 Approximate Inference Simon Keizer slides pt1
 slides pt2
Loopy Belief Propagation for Approximate Inference: An Empirical Study
Expectation Propagation for Approximate Bayesian Inference
 20.May.2013 Kernel Methods Zhuoran Wang slides Kernel Methods in Machine Learning
 27.May.2013Conditional Random Fields
 Nina Dethlefs slides An Introduction to Conditional Random Fields for Relational Learning
 03.Jun.2013  Inductive Logic Programming (postponed)
 Gudmund Grov  
 10.Jun.2013 Transfer Learning 2 Simon Keizer slides
 A Survey on Transfer Learning


 07.Oct.2013 Hierarchical Joint Learning for Natural Language Generation Nina Dethlefs  
 14.Oct.2013 Multilabel Classifier Systems Dimitra Gkatzia slides An extensive experimental comparison of methods for multi-label learning
 21.Oct.2013 Multi-Robot Strategy Learning Aris Valtazanos* slides Bayesian Interaction Shaping: Learning to Influence Strategic Interactions in Mixed Robotic Domains
 28.Oct.2013 Hierarchy Discovery in Reinforcement Learning Heriberto Cuayáhuitl slides Automatic Discovery and Transfer of Task Hierarchies in Reinforcement Learning
 04.Nov.2013 Linear Programming and its Application to MLE Tom Larkworthy  Applied Mathematical Programming (1977)  Chapter 9
 11.Nov.2013 Local Linear Weighted Projection Regression (postponed) Georgios Fagogenis  
 18.Nov.2013 Learning from Crowds Simon Keizer slides Learning from Crowds
 25.Nov.2013  Alexander Vezhnevets*  Large-scale Knowledge Transfer for Object Localization in ImageNet
 02.Dec.2013 Structured Prediction Zhuoran Wang slides Support vector machine learning for interdependent and structured output spaces
Learning Structured Prediction Models: A Large Margin Approach
 09.Dec.2013 Online Feature Learning  Arash Eshghi  Simultaneous Feature Selection and Parameter Optimization for Training of Dialogue Policy by Reinforcement Learning

Intro to ML + decision trees
Heriberto Cuayáhuitl
Mitchell: Chapter 3
The discipline of machine learning
Learning to fly
Review of probability and estimation
Zhuoran Wang
Bishop: Chapter 1 thru 1.2.3 and Chapter 2 thru 2.2
Bayesian Inference: An Introduction to Principles and Practice in Machine Learning
Bayesian Methods for Support Vector Machines: Evidence and Predictive Class Probabilities
My Slides
Naive Bayes
Kevin O'Connor
slidesGenerative and Discriminative Classifiers: Naive Bayes and logistic regression
Naive Bayes optimality: exploring the independence assumption
Naive Bayes models and EM, derivation of MLE estimates
Tackling poor assumptions of NB
Graphical models 1 (postponed one week)
Simon Keizer
Bishop: Chapter 8, Section 1-3
Recommended video lecture on graphical models by Zoubin Ghahramani
Graphical models 2 (postponed to 2013)
Simon Keizer
Bishop: Chapter 8, Section 4; ...
Active learning
Heriberto CuayáhuitlslidesActive Learning Literature Survey
Active Learning with SVMs Applied to Gene Expression Data for Cancer Classification
Active Learning for Spoken Language Understanding
Transparent Active Learning for Robots
Semi-supervised learning (Cotraining)Tom Larkworthy
Theory: Combining Labeled and Unlabeled Data with Co-Training
Practice: Toward an Architecture for Never-Ending Language Learning

Hidden Markov modelsNina Dethlefsslides
Bishop: Chapter 13
A Couple HMM for Audio-Visual Speech Recognition
Jurafsky and Martin: Chapter 6.
Reinforcement learning 1 (MDPs)
Ioannis Efstathiou
slidesMitchell: Chapter 13 (for details about the book please see below)
Reinforcement Learning: A Survey

Machine Learning Books:

Machine Learning Courses:

Machine Learning Tools:

Do you want to send a message to the reading group?

Just send an email to the following address: HW-ML-group (at) googlegroups (dot) com

Do you want to become a member of the group?

Just send an email to the following address: s (dot) keizer (at) hw (dot) ac (dot) uk

Organisers: Simon Keizer
Simon Keizer,
10 Jun 2013, 10:08
Heriberto Cuayahuitl,
3 Jun 2013, 10:34
Simon Keizer,
14 Oct 2013, 06:53
Heriberto Cuayahuitl,
14 Dec 2015, 08:34
Simon Keizer,
6 Nov 2015, 03:17