CS 159: Advanced Topics in Machine Learning: Structured Prediction

2016/2017 Spring Term (previous year)

Course Description

This course will cover a mixture of the following topics:

Course Details


Taehwan Kim           taehwan@caltech.edu
Yisong Yue               yyue@caltech.edu

Teaching Assistants

Hoang Le                 hmle@caltech.edu
Jialin Song               jssong@caltech.edu
Stephan Zheng        stzheng@caltech.edu

Office Hours

Datasets and Codes

Could be useful for final project.

Presentation Schedule

Note: schedule is subject to change.

Date Papers Presenters                     Materials
4/4/2017  Introduction & Administrivia 
Prababilistic Graphical Models 
Inference Methods
Taehwan Kim [slides]
  • An introduction to graphical models
  • Probabilistic inference in graphical models
  • 4/6/2017  Learning for Structured Prediction
    Structured Perceptrons & Structural SVMs
    Yisong Yue [slides]
  • Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with Perceptron Algorithms
  • Large Margin Methods for Structured and Interdependent Output Variables
  • 4/11/2017 Conditional Random Fields Milan Cvitkovic,
    Grant Van Horn
    Mentor: Taehwan Kim
  • Conditional random fields: Probabilistic models for segmenting and labeling sequence data
  • Conditional Random Fields for Object Recognition
  • Conditional Random Fields as Recurrent Neural Networks
  • Introduction to Conditional Random Field (optional: a comprehensive tutorial)
  • 4/13/2017 Message Passing &
    Linear Programing
    Hoang Le,
    Jialin Song
  • Probabilistic inference in graphical models
  • Chapter 8.2 and 8.4 in Graphical Models, Exponential Families, and Variational Inference
  • Fixing Max-Product: Convergent Message Passing Algorithms for MAP LP-Relaxations
  • 4/18/2017 Variational Inference Kun Ho (John) Kim,
    Albert Zhao
    Mentor: Taehwan Kim
  • Variational Inference: A Review for Statisticians
  • Crowdclustering
  • Tutorial on Variational Autoencoders (optional)
  • 4/20/2017 Sampling Methods Sara Beery,
    Natalie Bernat,
    Eric Zhan
    Mentor: Hoang Le
  • An Introduction to MCMC for Machine Learning
  • An MCMC-based Particle Filter for Tracking Multiple Interacting Targets
  • 4/25/2017 Hidden Markov Models (HMM) Gabriela Tavares,
    Juri Minxha
    Mentor: Taehwan Kim
  • Large margin hidden Markov models for automatic speech recognition
  • Hidden Markov Support Vector Machines
  • Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition
  • 4/27/2017 Topic Model Zach Wu,
    Kevin Yang
    Mentor: Taehwan Kim
  • Latent Dirichlet Allocation
  • Introduction to Probabilistic Topic Models
  • Topics over Time: A Non-Markov Continuous-Time Model of Topical Trends
  • 5/2/2017 Hierarchical / Extreme Classification Anish Thilagar,
    Ashwin Balakrishna
    Mentor: Jialin Song
  • Hierarchical document categorization with support vector machines
  • A Primal and Dual Sparse Approach to Extreme Classification
  • Logarithmic Time Online Multiclass prediction
  • 5/4/2017 Structured Perceptron Ruofei Shen,
    Xinghui Lu,
    Yue Lu,
    Ye Qiu
    Mentor: Yisong Yue
  • Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with Perceptron Algorithms
  • Distributed Training Strategies for the Structured Perceptron
  • 5/9/2017 Structured Random Forests Benyamin AllahGholiZadeh Haghi,
    Cody Han,
    Yury Tokpanov
    Mentor: Taehwan Kim
  • Structured Forests for Fast Edge Detection
  • Structured Class-Labels in Random Forests for Semantic Image Labelling
  • Decision Forests: A Unified Framework for Classification, Regression, Density Estimation, Manifold Learning and Semi-Supervised Learning (optional: a comprehensive tutorial)
  • 5/11/2017 Deep Structured Models #1: Graphical Models + Deep Learning Yongliang Zhang,
    Ruoqi Shen,
    Yu Su,
    Chen Liang
    Mentor: Hoang Le
  • Learning Deep Structured Models
  • Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
  • Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
  • 5/16/2017 Deep Structured Models #2: Deep (Convolutional) Neural Networks Qilin Li,
    Xinjie Lei,
    Hongnian Yu,
    DiJia Su
    Mentor: Yisong Yue
  • Learning Structured Output Representation using Deep Conditional Generative Models
  • Learning to Generate Chairs with Convolutional Neural Networks
  • Label-Free Supervision of Neural Networks with Physics and Domain Knowledge(optional)
  • 5/18/2017 Deep Structured Models #3: Recurrent Neural Networks Rohan Batra,
    Audrey Huang,
    Nand Kishore
    Mentor: Stephan Zheng
  • Sequence Level Training with Recurrent Neural Networks
  • Structure Inference Machines: Recurrent Neural Networks for Analyzing Relations in Group Activity Recognition
  • 5/23/2017 Sequence-to-sequence Model Albert Ge,
    Timothy Chou,
    Andrew Ding
    Mentor: Taehwan Kim
  • A Decision Tree Framework for Spatiotemporal Sequence Prediction
  • Sequence to Sequence Learning with Neural Networks
  • Neural machine translation by jointly learning to align and translate
  • 5/25/2017 Image Captioning and Generation From Text Jonathan Kenny,
    Tony Zhang,
    Jeremy Bernstein
    Mentor: Stephan Zheng
  • Attend and Tell: Neural Image Caption Generation with Visual Attention
  • Generating images from captions with attention
  • 5/30/2017 Active Learning for Structured Prediction Joey Hong,
    Rohan Doshi,
    Rohan Choudhury
    Mentor: Jialin Song
  • Online Structured Prediction via Coactive Learning
  • Latent Structured Active Learning
  • Topic and Reading List

    Presentation Signup Sheet

    Extended Reference Material (could be useful for picking final project)

    Note: some papers belong to multiple categories.

    Related Courses, Tutorials and Textbooks