Timezone: »

Sets and Partitions
Nicholas Monath · Manzil Zaheer · Andrew McCallum · Ari Kobren · Junier Oliva · Barnabas Poczos · Ruslan Salakhutdinov

Sat Dec 14 08:00 AM -- 06:00 PM (PST) @ West 215 + 216
Event URL: https://www.sets.parts »

Classic problems for which the input and/or output is set-valued are ubiquitous in machine learning. For example, multi-instance learning, estimating population statistics, and point cloud classification are all problem domains in which the input is set-valued. In multi-label classification the output is a set of labels, and in clustering, the output is a partition. New tasks that take sets as input are also rapidly emerging in a variety of application areas including: high energy physics, cosmology, crystallography, and art. As a natural means of succinctly capturing large collections of items, techniques for learning representations of sets and partitions have significant potential to enhance scalability, capture complex dependencies, and improve interpretability. The importance and potential of improved set processing has led to recent work on permutation invariant and equivariant representations (Ravanbakhsh et al, 2016; Zaheer et al, 2017; Ilse et al, 2018; Hartford et al, 2018; Lee et al, 2019, Cotter et al, 2019, Bloom-Reddy & Teh, 2019, and more) and continuous representations of set-based outputs and partitions (Tai and Lin, 2012; Belanger & McCallum, 2015; Wiseman et al, 2016; Caron et al, 2018; Zhang et al, 2019; Vikram et al 2019).

The goal of this workshop is to explore:
- Permutation invariant and equivariant representations; empirical performance, limitations, implications, inductive biases of proposed representations of sets and partitions, as well as rich models of interaction among set elements;
- Inference methods for predicting sets or clusterings; approaches based on gradient-descent, continuous representations, amenable to end-to-end optimization with other models;
- New applications of set and partition-based models.

The First Workshop on Sets and Partitions, to be held as a part of the NeurIPS 2019 conference, focuses on models for tasks with set-based inputs/outputs as well as models of partitions and novel clustering methodology. The workshop welcomes both methodological and theoretical contributions, and also new applications. Connections to related problems in optimization, algorithms, theory as well as investigations of learning approaches to set/partition problems are also highly relevant to the workshop. We invite both paper submissions and submissions of open problems. We hope that the workshops will inspire further progress in this important field.

Organizing Committee:
Andrew McCallum, UMass Amherst
Ruslan Salakhutdinov, CMU
Barnabas Poczos, CMU
Junier Oliva, UNC Chapel Hill
Manzil Zaheer, Google Research
Ari Kobren, UMass Amherst
Nicholas Monath, UMass Amherst
with senior advisory support from Alex Smola.

Invited Speakers:
Siamak Ravanbakhsh
Abhishek Khetan
Eunsu Kang
Amr Ahmed
Stefanie Jegelka

Sat 8:45 a.m. - 9:00 a.m.
Opening Remarks (Talk)
Manzil Zaheer, Nicholas Monath, Ari Kobren, Junier Oliva, Barnabas Poczos, Ruslan Salakhutdinov, Andrew McCallum
Sat 9:00 a.m. - 9:45 a.m.
Invited Talk - Stefanie Jegelka - Set Representations in Graph Neural Networks and Reasoning (Talk)
Stefanie Jegelka
Sat 9:45 a.m. - 10:30 a.m.

Poster Session 1 Paper Titles & Authors:

Deep Set Prediction Networks. Yan Zhang, Jonathon Hare, Adam Prügel-Bennett

Deep Hyperedges: a Framework for Transductive and Inductive Learning on Hypergraphs. Joshua Payne

FSPool: Learning Set Representations with Featurewise Sort Pooling. Yan Zhang, Jonathon Hare, Adam Prügel-Bennett

Deep Learning Features Through Dictionary Learning with Improved Clustering for Image Classification. Shengda Luo, Alex Po Leung, Haici Zhang

Globally Optimal Model-based Clustering via Mixed Integer Nonlinear Programming. Patrick Flaherty, Pitchaya Wiratchotisatian, Andrew C. Trapp

Sliding Window Algorithms for k-Clustering Problems. Michele Borassi, Alessandro Epasto, Silvio Lattanzi, Sergei Vassilvitski, Morteza Zadimoghaddam

Optimized Recommendations When Customers Select Multiple Products. Prasoon Patidar, Deeksha Sinha, Theja Tulabandhula

Manipulating Person Videos with Natural Language. Levent Karacan, Mehmet Gunel, Aykut Erdem, Erkut Erdem

Permutation Invariance and Relational Reasoning in Multi-Object Tracking. Fabian B. Fuchs, Adam R. Kosiorek, Li Sun, Oiwi Parker Jones, Ingmar Posner.

Clustering by Learning to Optimize Normalized Cuts. Azade Nazi, Will Hang, Anna Goldie, Sujith Ravi, Azalia Mirhoseini

Deformable Filter Convolution for Point Cloud Reasoning. Yuwen Xiong, Mengye Ren, Renjie Liao, Kelvin Wong, Raquel Urtasun

Learning Embeddings from Cancer Mutation Sets for Classification Tasks. Geoffroy Dubourg-Felonneau, Yasmeen Kussad, Dominic Kirkham, John Cassidy, Harry W Clifford

Exchangeable Generative Models with Flow Scans. Christopher M. Bender, Kevin O'Connor, Yang Li, Juan Jose Garcia, Manzil Zaheer, Junier Oliva

Conditional Invertible Flow for Point Cloud Generation. Stypulkowski Michal, Zamorski Maciej, Zieba Maciej, Chorowski Jan

Getting Topology and Point Cloud Generation to Mesh. Austin Dill, Chun-Liang Li, Songwei Ge, Eunsu Kang

Distributed Balanced Partitioning and Applications in Large-scale Load Balancing. Aaron Archer, Kevin Aydin, MohammadHossein Bateni, Vahab Mirrokni, Aaron Schild, Ray Yang, Richard Zhuang

Yan Zhang, Jonathon Hare, Adam Prugel-Bennett, Alex Leung, Patrick Flaherty, Pitchaya Wiratchotisatian, Alessandro Epasto, Silvio Lattanzi, Sergei Vassilvitskii, Morteza Zadimoghaddam, Theja Tulabandhula, Fabian Fuchs, Adam Kosiorek, Ingmar Posner, William Hang, Anna Goldie, Sujith Ravi, Azalia Mirhoseini, Yuwen Xiong, Mengye Ren, Renjie Liao, Raquel Urtasun, Haici Zhang, Michele Borassi, Shengda Luo, Andy Trapp, Geoffroy Dubourg-Felonneau, Yasmeen Kussad, Chris Bender, Manzil Zaheer, Junier Oliva, Michał Stypułkowski, Maciej Zieba, Austin Dill, Chun-Liang Li, Songwei Ge, Eunsu Kang, Oiwi Parker Jones, Kelvin Ka Wing Wong, Josh Payne, Yang Li, Azade Nazi, Erkut Erdem, Aykut Erdem, Kevin O'Connor, Juan J Garcia, Maciej Zamorski, Jan Chorowski, Deeksha Sinha, Harry Clifford, John W Cassidy
Sat 10:30 a.m. - 11:15 a.m.
Invited Talk - Siamak Ravanbakhsh - Equivariant Multilayer Perceptrons (Talk)
Siamak Ravanbakhsh
Sat 11:15 a.m. - 11:30 a.m.
Contributed Talk - Towards deep amortized clustering (Talk)
Juho Lee, Yoonho Lee, Yee Whye Teh
Sat 11:30 a.m. - 11:45 a.m.
Contributed Talk - Fair Hierarchical Clustering (Talk)
Sara Ahmadian, Alessandro Epasto, Marina Knittel, Ravi Kumar, Mohammad Mahdian, Philip Pham
Sat 11:45 a.m. - 12:30 p.m.
Invited Talk - Abhishek Khetan - Molecular geometries as point clouds: Learning physico-chemical properties using DeepSets (Talk)
Abhishek Khetan
Sat 12:30 p.m. - 2:00 p.m.
Lunch Break (on your own) (Break)
Sat 2:00 p.m. - 2:15 p.m.

Limitations of Deep Learning on Point Clouds Christian Bueno, Alan G. Hylton

Christian Bueno
Sat 2:15 p.m. - 2:30 p.m.
Contributed Talk - Chirality Nets: Exploiting Structure in Human Pose Regression (Talk)
Raymond Yeh, Yuan-Ting Hu, Alex Schwing
Sat 2:30 p.m. - 3:15 p.m.
Invited Talk - Eunsu Kang - Sets for Arts (Talk)
Eunsu Kang
Sat 3:15 p.m. - 4:15 p.m.

Poster Session 2 Paper Titles & Authors:

Towards deep amortized clustering. Juho Lee, Yoonho Lee, Yee Whye Teh

Chirality Nets: Exploiting Structure in Human Pose Regression. Raymond Yeh, Yuan-Ting Hu, Alexander Schwing

Fair Hierarchical Clustering. Sara Ahmadian, Alessandro Epasto, Marina Knittel, Ravi Kumar, Mohammad Mahdian, Philip Pham

Limitations of Deep Learning on Point Clouds. Christian Bueno, Alan G. Hylton

How Powerful Are Randomly Initialized Pointcloud Set Functions? Aditya Sanghi, Pradeep Kumar Jayaraman

On the Possibility of Rewarding Structure Learning Agents: Mutual Information on Linguistic Random Sets. Ignacio Arroyo-Fernández, Mauricio Carrasco-Ruiz, José Anibal Arias-Aguilar

Modelling Convolution as a Finite Set of Operations Through Transformation Semigroup Theory. Andrew Hryniowski, Alexander Wong

HCA-DBSCAN: HyperCube Accelerated Density Based Spatial Clustering for Applications with Noise. Vinayak Mathur, Jinesh Mehta, Sanjay Singh

Finding densest subgraph in probabilistically evolving graphs. Sara Ahmadian, Shahrzad Haddadan

Representation Learning with Multisets. Vasco Portilheiro

PairNets: Novel Fast Shallow Artificial Neural Networks on Partitioned Subspaces. Luna Zhang

Fair Correlation Clustering. Sara Ahmadian, Alessandro Epasto, Ravi Kumar, Mohammad Mahdian

Learning Maximally Predictive Prototypes in Multiple Instance Learning. Mert Yuksekgonul, Ozgur Emre Sivrikaya, Mustafa Gokce Baydogan

Deep Clustering using MMD Variational Autoencoder and Traditional Clustering Algorithms. Jhosimar Arias

Hypergraph Partitioning using Tensor Eigenvalue Decomposition. Deepak Maurya, Balaraman Ravindran, Shankar Narasimhan

Information Geometric Set Embeddings: From Sets to Distributions. Ke Sun, Frank Nielsen

Document Representations using Fine-Grained Topics. Justin Payan, Andrew McCallum

Juho Lee, Yoonho Lee, Yee Whye Teh, Raymond Yeh, Yuan-Ting Hu, Alex Schwing, Sara Ahmadian, Alessandro Epasto, Marina Knittel, Ravi Kumar, Mohammad Mahdian, Christian Bueno, Aditya Sanghi, Pradeep Kumar Jayaraman, Ignacio Arroyo-Fernández, Andrew Hryniowski, Vinayak Mathur, Sanjay Singh, Shahrzad Haddadan, Vasco Portilheiro, Luna Zhang, Mert Yuksekgonul, Jhosimar Arias Figueroa, Deepak Maurya, Balaraman Ravindran, Frank NIELSEN, Philip Pham, Justin Payan, Andrew McCallum, Jinesh Mehta, K Sun
Sat 4:15 p.m. - 5:00 p.m.
Invited Talk - Alexander J. Smola - Sets and symmetries (Talk)
Alex Smola
Sat 5:00 p.m. - 5:40 p.m.
Panel Discussion
Sat 5:40 p.m. - 5:45 p.m.
Closing Remarks (Talk)

Author Information

Nicholas Monath (University of Massachusetts Amherst)
Manzil Zaheer (Google)
Andrew McCallum (UMass Amherst)
Ari Kobren (UMass Amherst)
Junier Oliva (UNC - Chapel Hill)
Barnabas Poczos (Carnegie Mellon University)
Ruslan Salakhutdinov (Carnegie Mellon University)

More from the Same Authors