Schedule for Spring 2015

Jan 12

• Planning meeting
  • Following previous semester, this semester, we will read more about some of the ML techniques that people use for their work. They will lead the discussion on that particular topic.

Jan 19

• Embedding Models (Michael)
  • WSABIE paper
    • http://www.thespermwhale.com/jaseweston/papers/wsabie-ijcai.pdf

Jan 26

• Graph Embeddings (Siva)
  • Open question answering with weakly supervised embedding models (Bordes et al. 2014)
    • http://arxiv.org/pdf/1404.4326.pdf
  • Question Answering with Subgraph Embeddings
    
    • http://aclweb.org/anthology/D/D14/D14-1067.pdf

Feb 2

• Compositional Vector Grammar : Overview (Shashi)
  • http://nlp.stanford.edu/pubs/SocherBauerManningNg_ACL2013.pdf

Feb 9

• Compositional Vector Grammar : CRF auto encoders (Daniel)
  • http://papers.nips.cc/paper/5344-conditional-random-field-autoencoders-for-unsupervised-structured-prediction.pdf

Mar 02

• Language+Vision I (Spandana)
  • http://cs.stanford.edu/people/karpathy/deepimagesent/devisagen_arxiv.pdf

Mar 09

• Language+Vision II (XinXing )
  • Google http://arxiv.org/pdf/1411.4555.pdf?
  • Baidu/UCLA http://arxiv.org/pdf/1410.1090v1.pdf

Mar 16

• Graph Theory I (Philip)
  • Identifying Sets of Key Players in a Social Network.
    • http://link.springer.com/article/10.1007%2Fs10588-006-7084-x

Mar 23

• Graph Theory II / Inference on Graphs (Philip)
  • Rolenet: Movie Analysis from the Perspective of Social Networks.
    • http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4757440&tag=1

Mar 30

• Topic Modeling I (Lea)
  • General Introduction & Theoretical Bounds for LDA: Understanding the Limiting Factors of Topic Modeling via Posterior Contraction Analysis. (Tang et al, 2014)
    • http://www-personal.umich.edu/~qmei/pub/icml2014-tang.pdf

Apr 6

• Cancelled (Easter Monday)

Apr 13

• Topic Modeling II ( Jari)
  • Hyperparameter optimization: Rethinking LDA: Why priors matter. (Wallach et al, 2009)
    • http://works.bepress.com/cgi/viewcontent.cgi?article=1018&context=hanna_wallach
  • Evaluating LDA: Evaluation methods for topic models. (Wallach et al, 2009)
    • http://www.era.lib.ed.ac.uk/bitstream/1842/4587/1/MurrayI_Evaluation%20Methods%20for.pdf

Apr 20

• Topic Modeling III ( Annie)
  • A Sense-Topic Model for Word Sense Induction with Unsupervised Data Enrichment
    • https://tacl2013.cs.columbia.edu/ojs/index.php/tacl/article/view/485/107

Apr 27

• Topic Modeling IV ( Jari)
  • Relational topic models: Relational topic models for document networks. (Chang et al, 2009)
    • http://machinelearning.wustl.edu/mlpapers/paper_files/AISTATS09_ChangB.pdf
  • Scoped LDA for multi-document summarization: Exploring content models for multi-document summarization. (Haghighi et al, 2009)
    • http://www.aclweb.org/anthology/N09-1041

May 4

• Tools Session: General (Siva, Nathan, Bharat)
  • Latex tricks, bash commands, pdf tools, ML tools

May 11

• Tolls Session: Neural Networks (Daniel)
  • Theano for Neural Networks

Misc. other topics

• Neural Nets : Hessian-free optimization (Daniel)
• Neural Nets : Bayesian Models for training (Daniel)
• Sequence models through Neural Networks: Theano (Daniel)
• Un-supervised grammar induction
• Unsupervised sequence segmentation : Bayesian vs. Neural Nets (Lea & Daniel)
• Shay's Bayesian NLP book
• Discriminative Models : Perceptron (Bharat)
• Discriminative Models : Stochastic Gradient Descent

Schedule for Monsoon 2014

Sep 18

• Planning meeting
  • Following previous semester, this semester, we will read more about some of the ML techniques that people use for their work. They will lead the discussion on that particular topic.

Sep 25

• Paper discussion (Lea)
  • Demystifying Information-Theoretic Clustering (Ver Steeg et al, 2014)
    • http://jmlr.org/proceedings/papers/v32/steeg14.pdf

Sep 29

• no meeting

Oct 6

• Unsupervised Grammar induction / Adaptor Grammars (Introductory paper : Annie)
  • Adaptor Grammars: A Framework for Specifying Compositional Nonparametric Bayesian Models
    • http://homepages.inf.ed.ac.uk/sgwater/papers/nips07-adaptor.pdf
  • Slides:
    • http://web.science.mq.edu.au/~mjohnson/papers/Johnson-learning-rules-10-talk.pdf

Oct 13

• Adaptor Grammars (Application Paper : Annie)
  • Modeling Perspective using Adaptor Grammars (Hardisty et al, 2010)
    • http://www.cs.umd.edu/sites/default/files/scholarly_papers/Hardisty_1.pdf

Oct 20

• Tree Substitution Grammars (Miltos)
  • http://homepages.inf.ed.ac.uk/sgwater/papers/naacl09_cohn.pdf

Oct 27

• Learning Word Embeddings (Kristian)
  • http://emnlp2014.org/papers/pdf/EMNLP2014162.pdf

Nov 3

• no meeting

Nov 10

• Neural Language Models (Miltos)
  • A fast and simple algorithm for training neural probabilistic language models
    • http://arxiv.org/pdf/1206.6426.pdf

Nov 17

• Word Embeddings : Application (Herman)
  • Semi-Supervised Recursive Autoencoders for Predicting Sentiment Distributions
    • http://nlp.stanford.edu/pubs/SocherPenningtonHuangNgManning_EMNLP2011.pdf
    • critique : http://uk.arxiv.org/abs/1301.2811

Nov 24

• Long Short Term Memory (LSTM) Neural Nets (Daniel and XingXing )
• lstm_intro.pdf: Slides for the LSTM intro

Dec 1

• Long Short Term Memory (LSTM) Neural Nets (Daniel and XingXing )

Dec 8

Misc. other topics

• Perceptron
• "Learning to Search"
• (Incremental) (Bayesian) learning in high-dimensional space
• (Introductory) Topic Modeling / Gibbs Sampling
• MERT/MIRA
• Analyzing the Errors of Unsupervised Learning (Liang and Klein, 2008)
• Graph Theory
• Inference on Graphs (Philip)

Previous Meetings

• Visit Previous meetings for a list of previous meetings

Useful Information

• Visit Useful Information for information about useful tools and paper recommendations from previous meetings

Other Reading Groups

• Cambridge http://www.wiki.cl.cam.ac.uk/rowiki/NaturalLanguage/ReadingGroup
• Toronto http://learning.cs.toronto.edu/mlreading.html