New England Machine Learning Day 2015

About

The fourth annual New England Machine Learning Day will be Monday, May 18, 2015, at Microsoft Research New England, One Memorial Drive, Cambridge, MA 02142. The event will bring together local academics and researchers in machine learning and its applications. There will be a lively poster session during lunch.

Related Events:

Agenda

9:50 – 10:00
Opening remarks

10:00 – 10:30, Tamara Broderick, MIT
Statistical and computational trade-offs from Bayesian approaches to unsupervised learning

The flexibility, modularity, and coherent uncertainty estimates provided by Bayesian posterior inference have made this approach indispensable in a variety of domains. Since posteriors for many problems of interest cannot be calculated exactly, much work has focused on delivering accurate posterior approximations—though the computational cost of these approximations can sometimes be prohibitive, particularly in a modern, large-data context. Focusing on unsupervised learning problems, we illustrate in a series of vignettes how we can trade off some typical Bayesian desiderata for computational gains and vice versa. On one end of the spectrum, we sacrifice learning uncertainty to deliver fast, flexible methods for point estimates. In particular, we consider taking limits of Bayesian posteriors to obtain novel K-means-like objective functions as well as scalable, distributed algorithms. On the other end, we consider mean-field variational Bayes (MFVB), a popular and fast posterior approximation method that is known to provide poor estimates of parameter covariance. We develop an augmentation to MFVB that delivers accurate estimates of posterior uncertainty for model parameters.

10:35 – 11:05, David Jensen, University of Massachusetts Amherst
Why ML Needs Causality

Machine learning has risen to extraordinary success and prominence primarily by representing and reasoning about statistical associations. Apparently, we can perform a great many useful tasks without having to concern ourselves with representing and reasoning about causality. In this talk, I will explain the key differences between associational and causal models and describe some of the recent technical developments in causal inference. I will explain why many existing machine learning tasks are really causal reasoning in disguise, why an increasing number of machine learning tasks will explicitly require causal models, and why researchers and practitioners who understand causal reasoning will succeed where others fail.

11:10 – 11:40, Tim Kraska, Brown
Tupleware: Redefining Modern Analytics

There is a fundamental discrepancy between the targeted and actual users of current analytics frameworks. Most systems are designed for the data and infrastructure of the Googles and Facebooks of the world—petabytes of data distributed across large cloud deployments consisting of thousands of cheap commodity machines. Yet, the vast majority of companies operate or rent clusters in the range of a few dozen nodes and analyze relatively small data sets of up to a few terabytes. Targeting these users fundamentally changes the way we should build analytics systems.In this talk, Tim will present Tupleware, a new system developed at Brown University specifically aimed at the challenges faced by the typical user. The main difference of Tupleware to other frameworks is, that it automatically compiles complex machine learning workflows into highly efficient distributed programs instead of interpreting the workflows at run-time. Our initial experiments show, that Tupleware is 30x – 300x faster than Spark and up to 6000x faster than Hadoop for common machine learning algorithms.

11:40 – 1:45
Lunch and posters

1:45 – 2:15, Suvrit Sra, MIT
Geometric optimization in machine learning

2:20 – 2:50, Jennifer Listgarten, Microsoft Research
Genome and Epigenome-Wide Association Studies

Understanding the genetic underpinnings of disease is important for screening, treatment, drug development, and basic biological insight. Genome-wide associations, wherein individual or sets of genetic markers are systematically scanned for association with disease are one window into disease processes. Naively, these associations can be found by use of a simple statistical test. However, a wide variety of confounders lie hidden in the data, leading to both spurious associations and missed associations if not properly addressed. These confounding factors include population structure, family relatedness, cell type heterogeneity, and environmental confounders. I will discuss the state-of-the art statistical approaches based on linear mixed models for conducting these analyses, in which the confounders are automatically deduced, and then corrected for, by the data and model.

2:50 – 3:20
Coffee break

3:20 – 3:50, Krzysztof Gajos, Harvard
Interactive Intelligent Systems

3:55 – 4:25, Jennifer Dy, Northeastern
Learning in Complex Data

Machine learning as a field has become more and more important due to the ubiquity of data collection in various disciplines. Coupled with this data collection is the hope that new discoveries or knowledge can be learned from data. My research spans both fundamental research in machine learning and the application of those methods to biomedical imaging, health, science and engineering. Multi-disciplinary research is instrumental to the growth of the various areas involved. Real problems provide rich sources of complex data that challenges the state of the art in machine learning; at the same time, machine learning algorithms provide additional resources to scientists in making non-trivial contributions to their field of research. In this talk, I present two examples of innovation in machine learning motivated by challenges from complex medical domains.Chronic Obstructive Pulmonary Disease (COPD) is a lung disease characterized by airflow limitation due to noxious particles or gases (e.g., cigarette smoke). COPD is known to be a heterogeneous disease with genetic factors predisposing individuals to varying levels of disease severity as a function of exposure. An improved understanding of this heterogeneity should lead to better stratification of patients for prognosis and personalized treatments. However, standard clustering algorithms are limited because they do not take into account the interplay of the different features. We introduce a transformative way of looking at subtyping/clustering by recasting it in terms of discovering associations of individuals to disease trajectories (i.e., grouping individuals based on their similarity in how their health changes as a response to environmental and/or disease causing variables). This led us to the development of a nonparametric Bayesian model of mixture of disease trajectories.

Skin cancer is one of the most common types of cancer. Standard clinical screening and diagnosis is performed using dermoscopy and visual examination. When an abnormal lesion is suspected, biopsy is carried out which is invasive, painful and leaves a scar. Our goal is to develop novel machine learning and image analysis algorithms that aid physicians in the early detection of skin cancer from 3D reflectance confocal microscopy images in vivo without resorting to biopsy. A key feature of interest is the dermal-epidermal junction which separates the dermis and epidermis layers of the skin. However, automated segmentation is challenging due to poor image contrast and high variability of objects in biology and medicine. This led us to the design of a novel generative probabilistic latent shape spatial Poisson process model that can take into account the uncertainty in number of objects, location, shape, and appearance. We develop a Gibbs sampler that addresses the variation in model order and nonconjugacy that arise in this setting.

4:30 – 5:00, Ankur Moitra, MIT
Beyond Matrix Completion

Here we study some of the statistical and algorithmic problems that arise in recommendation systems. We will be interested in what happens when we move beyond the matrix setting, to work with higher order objects — namely, tensors. To what extent does inference over more complex objects yield better predictions, but at the expense of the running time? We will explore the computational vs. statistical tradeoffs for some basic problems about recovering approximately low rank tensors from few observations, and will show that our algorithms are nearly optimal among all polynomial time algorithms, under natural complexity-theoretic assumptions.

This is based on joint work with Boaz Barak.

Posters


Poster Title 
Presenting Author/
Complete List of Authors
Supersparse Linear Integer Models for Optimized Medical Scoring Systems Berk Ustun, MIT/Berk Ustun, Cynthia Rudin
A Robotic Grasping System With Bandit-Based Adaptation

 

John Oberlin, Brown University/John Oberlin, Stefanie Tellex
Reliable and scalable variational inference for the hierarchical Dirichlet process

 

Michael C. Hughes, Brown University/Michael C. Hughes, Dae Il Kim, Erik B. Sudderth
Learning Propositional Functions in Large State Spaces for Planning and Reinforcement Learning

 

David Hershkowitz, Brown University, Computer Science Department (undergrad)/D. Ellis Hershkowitz, James MacGlashan, Stefanie Tellex
Optimization as Estimation with Gaussian Process Bandits Zi Wang, CSAIL, MIT/Zi Wang, Stefanie Jegelka, Leslie Pack Kaelbling, Tomás Lozano-Pérez

 

Estimating the Partition Function by Discriminance Sampling

 

Qiang Liu, MIT/Qiang Liu, Jian Peng, Alexander Ihler, John Fisher
Preserving Modes and Messages via Diverse Particle Selection Jason L. Pacheco, Brown/Jason L. Pacheco, Silvia Zuffi, Michael J. Black, Erik B. Sudderth

 

Data Mining Methods for Analyzing and Modeling Spatiotemporal Data

 

Dawei Wang, UMass Boston/Dawei Wang, Wei Ding, Kui Yu
Bayesian Or?s of And?s for Interpretable Classification, with Application to Context-Aware Recommender Systems Tong Wang, MIT/Tong Wang, Cynthia Rudin, Finale Doshi-Velez, Yimin Liu, Erica Klampfl, Perry MacNeille

 

Cheap Bandits Manjesh Hanawal, BU/Manjesh K Hanawal, Venkatesh Saligrama

 

Modeling and Prediction of Diabetes-Related Hospitalizations Using Electronic Health Records

 

Tingting Xu, BU/Theodora S. Brisimi, Tingting Xu, Ioannis Ch. Paschalidis

 

Adding Expert Knowledge in Sparse Learning of High Dimensional Diffusion Data in Traumatic Brain Injury

 

Matineh Shaker, Northeastern/Matineh Shaker, Deniz Erdogmus, Jennifer Dy, Sylvain Bouix
Machine Learning and Dynamic Programming Algorithms for Motion Planning and Control

 

Oktay Arslan, Georgia Tech/Oktay Arslan, Panagiotis Tsiotras
Vector-Valued Property Elicitation

 

Rafael Frongillo, Harvard/Rafael Frongillo, Ian Kash
Gradient-based Hyperparameter Optimization through Reversible Learning

 

David Duvenaud, Harvard/Dougal Maclaurin, David Duvenaud, Ryan P. Adams
Learning Heterogeneous Progression Patterns of Alzheimer’s Disease with Clustered Hidden Markov Model

 

Chenhui Hu, Harvard/Chenhui Hu, Xiaoxiao Li, Finale Doshi-Velez, Xue Hua, Paul Thompson, Georges Fakhri, Quanzheng Li
Exploring Collections with Interactive Spatial Organization

 

Kenneth C. Arnold, Harvard/Kenneth C. Arnold, Krzysztof Z. Gajos
ChordRipple: More Creative Chord Recommendations with chord2vec

 

Anna Huang, Harvard/Cheng-Zhi Anna Huang, David Duvenaud, Krzysztof Z. Gajos
Stability and optimality in stochastic gradient descent

 

Dustin Tran, Harvard/Dustin Tran, Panos Toulis, Edoardo M. Airoldi
Multi-Level Dirichlet Priors for Modelling Topical Variations across Textual Regions Kriste Krstovski, Harvard-Smithsonian Center for Astrophysics/Kriste Krstovski, David A. Smith, Michael J. Kurtz

 

Scalable Bayesian Nonparametric Policy Learning in Dec-POMDPs Miao Liu, MIT LIDS/Miao Liu, Chris Amato, Jonathan P. How

 

Measuring semantics assumptions

 

Andrea Censi, MIT/Andrea Censi
SQUARE: A Benchmark for Consensus Algorithms in Crowdsourcing

 

Matt Lease, University of Texas at Austin/Matthew Lease, Aashish Sheshadri
A Nearly-Linear Time Framework for Graph-Structured Sparsity Chinmay Hegde, MIT/Chinmay Hegde, Piotr Indyk, Ludwig Schmidt

 

Sample-Optimal Density Estimation in Nearly-Linear Time Jerry Li, MIT/Jayadev Acharya, Ilias Diakonikolas, Jerry Li, Ludwig Schmidt

 

A Nearly Optimal and Agnostic Algorithm for Properly Learning a Mixture of k Gaussians, for any Constant k

 

Ludwig Schmidt, MIT/Jerry Li, Ludwig Schmidt
Nonparametric Bayesian Inference of Strategy in Infinitely Repeated Games Max Kleiman-Weiner, MIT/Max Kleiman-Weiner, Penghui Zhou, Joshua B. Tenenbaum

 

Perturbation Training for Human-Robot Teams

 

Ramya Ramakrishnan, MIT/Ramya Ramakrishnan, Chongjie Zhang, Julie A. Shah
Fixed-point algorithms for learning Determinantal Point Processes

 

Zelda Mariet, MIT/Zelda Mariet, Suvrit Sra

 

Object-based World Modeling with Dependent Dirichlet Process Mixtures

 

Lawson Wong, MIT CSAIL/Lawson L.S. Wong, Thanard Kurutach, Leslie Pack Kaelbling, Tomás Lozano-Pérez
Handwritten Tamil Character Recognition using a Convolutional Neural Network

 

Misha Sra, MIT Media lab/Prashanth Vijayaraghavan, Misha Sra
Computing Sparse/Low-rank/Structured Optimization Solutions using an Extension of the Frank-Wolfe Method, with Application to Matrix Completion

 

Paul Grigas, MIT Operations Research Center/Robert M. Freund, Paul Grigas, Rahul Mazumder
A Sparse Combined Regression-Classification Formulation for Learning a Physiological Alternative to Clinical Post-Traumatic Stress Disorder Scores

 

Sarah Brown, Northeastern University/Sarah Brown, Rami Mangoubi, Andrea Webb, Jennifer Dy
RSVP Keyboard A Brain Computer Interface for AAC Technology

 

Paula Gonzalez, Northeastern University/U. Orhan, M. Moghadamfalahi, P. Gonzalez-Navarro, B. Girvent, A. Ahani, M. Haghighi, B. Peters, A. Mooney, A. Fowler, K. Gorman, S. Bedrick, B. Oken, M. Akcakaya, M. Fried-Oken, D. Erdogmus

 

Personality Prediction in Heterogeneous Social Networks with Incomplete Attributes

 

Yuan Zhong, Northeastern University/Yuan Zhong, Yizhou Sun, Wen Zhong, Rui Dong, Yupeng Gu
Clustering and Ranking in Heterogeneous Information Networks via Gamma-Poisson Model

 

Junxiang Chen, Northeastern University/Junxiang Chen, Wei Dai, Yizhou Sun, Jennifer Dy
Efficient exploration of large molecular spaces with artificial neural networks

 

Edward Pyzer-Knapp, Harvard/Jos‚ Miguel Hern ndez-Lobato, Edward Pyzer-Knapp, Ryan Adams, Al n Aspuru-Guzik

 

Sparse Variational Inference for Generalized Gaussian Process Models Rishit Seth, Tufts University/Rishit Sheth, Bernie Wang, Roni Khardon
Method of Moments Learning for Left to Right Hidden Markov Models

 

Cem Sbakan, UIUC CS department/Cem Subakan, Johannes Traa, Paris Smaragdis, Daniel Hsu
Concurrent and Incremental Transfer Learning in a Network of Reinforcement Learning Agents Dan Garant, University of Massachusetts-Amherst/Dan Garant, Bruno Castro da Silva, Victor Lesser, Chongjie Zhang
Alternative approaches to discovering causality with additive noise models

 

Kaleigh Clary, University of Massachusetts-Amherst/ Kaleigh Clary, David Jensen

 

Cancer Subtype Discovery Using Machine Learning Henry Lo, University of Massachusetts-Boston/Henry Lo, Melissa Cruz, Dawei Wang, Wei Ding, Marieke Kuijjer, Heather Selby, John Quackenbush
Inferring Polyadic Events With Poisson Tensor Factorization

 

Aaron Schein, University of Massachusetts-Amherst/Aaron Schein, John Paisley, David M. Blei, Hanna M. Wallach

 

Towards understanding the Boundary Forest Algorithm Charles Mathy, Disney Research/Jose Bento, Charles Mathy, Dan Schmidt
Predictive Entropy Search for Bayesian Optimization with Unknown Constraints

 

Michael Gelbart, Harvard/Jos‚ Miguel Hern ndez-Lobato, Michael Gelbart, Matt Hoffman, Ryan Adams, Zoubin Ghahramani

 

Model Selection by Linear Programming

 

Tolga Bolukbasi, BU/Joseph Wang, Tolga Bolukbasi, Kirill Trapeznikov, Venkatesh Saligrama

 

Bag of Words Approach to Activity Classification Kevin Amaral, Northeastern/Kevin M. Amaral, Wei Ding, Scott E. Crouter, Ping Chen

 

NeurOS and NeuroBlocks:ÿ A Neural/Cognitive Operating System and Building Blocks Lee Scheffler, Cognitivity/Lee Scheffler
Feature-Budgeted Random Forest Feng Nan, Boston University/Feng Nan, Joseph Wang, Venkatesh Saligrama

 

For any questions, please contact us here.

Committee

Carla Brodley, Northeastern
Finale Doshi-Velez, Harvard
Stefanie Jegelka, MIT
Adam Tauman Kalai, Microsoft Research

The steering committee that selects the organizers of ML Day each year consists of Ryan Adams, Sham Kakade, Adam Tauman Kalai, and Joshua Tenenbaum.