Publications

Simulating Human Gaze with Neural Visual Attention
Leo Schwinn
Doina Precup
Bjoern Eskofier
Dario Zanca
2022-10-19
NeurIPS.cc/2022/Workshop/GMML (oral)
doi.org
openreview.net
The Paradox of Choice: On the Role of Attention in Hierarchical Reinforcement Learning
Andrei Cristian Nica
Khimya Khetarpal
Doina Precup
Decision-making AI agents are often faced with two important challenges: the depth of the planning horizon, and the branching factor due to … (see more)having many choices. Hierarchical reinforcement learning methods aim to solve the first problem, by providing shortcuts that skip over multiple time steps. To cope with the breadth, it is desirable to restrict the agent's attention at each step to a reasonable number of possible choices. The concept of affordances (Gibson, 1977) suggests that only certain actions are feasible in certain states. In this work, we first characterize "affordances" as a "hard" attention mechanism that strictly limits the available choices of temporally extended options. We then investigate the role of hard versus soft attention in training data collection, abstract value learning in long-horizon tasks, and handling a growing number of choices. To this end, we present an online, model-free algorithm to learn affordances that can be used to further learn subgoal options. Finally, we identify and empirically demonstrate the settings in which the "paradox of choice" arises, i.e. when having fewer but more meaningful choices improves the learning speed and performance of a reinforcement learning agent.
2022-10-19
NeurIPS.cc/2022/Workshop/Attention (poster)
openreview.net
openreview.net
Timeliness of reporting of SARS-CoV-2 seroprevalence results and their utility for infectious disease surveillance
Claire Donnici
Natasha Ilincic
Christian Cao
Caseng Zhang
Gabriel Deveaux
David A. Clifton
David L Buckeridge
Niklas Bobrovitz
Rahul K. Arora
2022-10-19
Epidemics (published)
doi.org
Unmasking the Lottery Ticket Hypothesis: Efficient Adaptive Pruning for Finding Winning Tickets
Mansheej Paul
Feng Chen
Brett W. Larsen
Jonathan Frankle
Surya Ganguli
Gintare Karolina Dziugaite
Modern deep learning involves training costly, highly overparameterized networks, thus motivating the search for sparser networks that requi… (see more)re less compute and memory but can still be trained to the same accuracy as the full network (i.e. matching). Iterative magnitude pruning (IMP) is a state of the art algorithm that can find such highly sparse matching subnetworks, known as winning tickets, that can be retrained from initialization or an early training stage. IMP operates by iterative cycles of training, masking a fraction of smallest magnitude weights, rewinding unmasked weights back to an early training point, and repeating. Despite its simplicity, the underlying principles for when and how IMP finds winning tickets remain elusive. In particular, what useful information does an IMP mask found at the end of training convey to a rewound network near the beginning of training? We find that—at higher sparsities—pairs of pruned networks at successive pruning iterations are connected by a linear path with zero error barrier if and only if they are matching. This indicates that masks found at the end of training encodes information about the identity of an axial subspace that intersects a desired linearly connected mode of a matching sublevel set. We leverage this observation to design a simple adaptive pruning heuristic for speeding up the discovery of winning tickets and achieve a 30% reduction in computation time on CIFAR-100. These results make progress toward demystifying the existence of winning tickets with an eye towards enabling the development of more efficient pruning algorithms.
2022-10-19
NeurIPS.cc/2022/Workshop/HITY (accepted)
openreview.net
openreview.net
Causal inference from text: A commentary
Dhanya Sridhar
David Blei
2022-10-18
Science Advances (published)
doi.org
Aligning MAGMA by Few-Shot Learning and Finetuning
Jean-Charles Layoun
Alexis Roger
Irina Rish
2022-10-17
ArXiv (preprint)
doi.org
arxiv.org
Generalizing in the Real World with Representation Learning
Tegan Maharaj
2022-10-17
ArXiv (preprint)
doi.org
arxiv.org
Social isolation and the brain in the pandemic era
Danilo Bzdok
Robin I. M. Dunbar
2022-10-17
Nature Human Behaviour (published)
doi.org
Adapting Triplet Importance of Implicit Feedback for Personalized Recommendation
Haolun Wu
Chen Ma
Yingxue Zhang
Xue Liu
Ruiming Tang
Mark J. Coates
2022-10-16
Proceedings of the 31st ACM International Conference on Information & Knowledge Management (published)
doi.org
arxiv.org
Better Modeling the Programming World with Code Concept Graphs-augmented Multi-modal Learning
Martin Weyssow
Houari Sahraoui
Bang Liu
The progress made in code modeling has been tremendous in recent years thanks to the design of natural language processing learning approach… (see more)es based on state-of-the-art model architectures. Nevertheless, we believe that the current state-of-the-art does not focus enough on the full potential that data may bring to a learning process in software engineering. Our vision articulates on the idea of leveraging multi-modal learning approaches to modeling the programming world. In this paper, we investigate one of the underlying idea of our vision whose objective based on concept graphs of identifiers aims at leveraging high-level relationships between domain concepts manipulated through particular language constructs. In particular, we propose to enhance an existing pretrained language model of code by joint-learning it with a graph neural network based on our concept graphs. We conducted a preliminary evaluation that shows gain of effectiveness of the models for code search using a simple joint-learning method and prompts us to further investigate our research vision.
2022-10-16
Proceedings of the ACM/IEEE 44th International Conference on Software Engineering: New Ideas and Emerging Results (published)
doi.org
arxiv.org
Identification of out-of-distribution cases of CNN using class-based surprise adequacy
Mira Marhaba
Ettore Merlo
Foutse Khomh
Giuliano Antoniol
Machine learning is vulnerable to possible incorrect classification of cases that are out of the distribution observed during training and c… (see more)alibration
2022-10-16
Proceedings of the 1st International Conference on AI Engineering: Software Engineering for AI (published)
doi.org
OptEmbed: Learning Optimal Embedding Table for Click-through Rate Prediction
Fuyuan Lyu
Xing Tang
Hao Zhu
Huifeng Guo
Yingxue Zhang
Ruiming Tang
Xue Liu
Click-through rate (CTR) prediction model usually consists of three components: embedding table, feature interaction layer, and classifier. … (see more)Learning embedding table plays a fundamental role in CTR prediction from the view of the model performance and memory usage. The embedding table is a two-dimensional tensor, with its axes indicating the number of feature values and the embedding dimension, respectively. To learn an efficient and effective embedding table, recent works either assign various embedding dimensions for feature fields and reduce the number of embeddings respectively or mask the embedding table parameters. However, all these existing works cannot get an optimal embedding table. On the one hand, various embedding dimensions still require a large amount of memory due to the vast number of features in the dataset. On the other hand, decreasing the number of embeddings usually suffers from performance degradation, which is intolerable in CTR prediction. Finally, pruning embedding parameters will lead to a sparse embedding table, which is hard to be deployed. To this end, we propose an optimal embedding table learning framework OptEmbed, which provides a practical and general method to find an optimal embedding table for various base CTR models. Specifically, we propose pruning the redundant embeddings regarding corresponding features' importance by learnable pruning thresholds. Furthermore, we consider assigning various embedding dimensions as one single candidate architecture. To efficiently search the optimal embedding dimensions, we design a uniform embedding dimension sampling scheme to equally train all candidate architectures, meaning architecture-related parameters and learnable thresholds are trained simultaneously in one supernet. We then propose an evolution search method based on the supernet to find the optimal embedding dimensions for each field. Experiments on public datasets show that OptEmbed can learn a compact embedding table which can further improve the model performance.
2022-10-16
Proceedings of the 31st ACM International Conference on Information & Knowledge Management (published)
doi.org
arxiv.org