Publications (24)
Deep RL With Information Constrained Policies: Generalization in Continuous Control
Tailia Malloy, Chris R. Sims, Tim Klinger +3
Biological agents learn and act intelligently in spite of a highly limited capacity to process and store information. Many real-world problems involve continuous control, which rep…
Routing Networks: Adaptive Selection of Non-linear Functions for Multi-Task Learning
Clemens Rosenbaum, Tim Klinger, Matthew Riemer
Multi-task learning (MTL) with neural networks leverages commonalities in tasks to improve performance, but often suffers from task interference which reduces the benefits of trans…
Transformers Learn Faster with Semantic Focus
Parikshit Ram, Kenneth L. Clarkson, Tim Klinger +2
Various forms of sparse attention have been explored to mitigate the quadratic computational and memory cost of the attention mechanism in transformers. We study sparse transformer…
Beyond Sliding Windows: Learning to Manage Memory in Non-Markovian Environments
Geraud Nangue Tasse, Matthew Riemer, Benjamin Rosman +1
Recent success in developing increasingly general purpose agents based on sequence models has led to increased focus on the problem of deploying computationally limited agents with…
Laziness Is a Virtue When It Comes to Compositionality in Neural Semantic Parsing
Maxwell Crouse, Pavan Kapanipathi, Subhajit Chaudhury +4
Nearly all general-purpose neural semantic parsers generate logical forms in a strictly top-down autoregressive fashion. Though such systems have achieved impressive results across…
Routing Networks and the Challenges of Modular and Compositional Computation
Clemens Rosenbaum, Ignacio Cases, Matthew Riemer +1
Compositionality is a key strategy for addressing combinatorial complexity and the curse of dimensionality. Recent work has shown that compositional solutions can be learned and of…
Learning interpretable positional encodings in transformers depends on initialization
Takuya Ito, Luca Cocchi, Tim Klinger +3
In transformers, the positional encoding (PE) provides essential information that distinguishes the position and order amongst tokens in a sequence. Most prior investigations of PE…
Hierarchical Reinforcement Learning with AI Planning Models
Junkyu Lee, Michael Katz, Don Joven Agravante +4
Two common approaches to sequential decision-making are AI planning (AIP) and reinforcement learning (RL). Each has strengths and weaknesses. AIP is interpretable, easy to integrat…
Logical Rule Induction and Theory Learning Using Neural Theorem Proving
Andres Campero, Aldo Pareja, Tim Klinger +2
A hallmark of human cognition is the ability to continually acquire and distill observations of the world into meaningful, predictive theories. In this paper we present a new mecha…
Scalable Recollections for Continual Lifelong Learning
Matthew Riemer, Tim Klinger, Djallel Bouneffouf +1
Given the recent success of Deep Learning applied to a variety of single tasks, it is natural to consider more human-realistic settings. Perhaps the most difficult of these setting…
What makes Models Compositional? A Theoretical View: With Supplement
Parikshit Ram, Tim Klinger, Alexander G. Gray
Compositionality is thought to be a key component of language, and various compositional benchmarks have been developed to empirically probe the compositional generalization of exi…
EXPLORER: Exploration-guided Reasoning for Textual Reinforcement Learning
Kinjal Basu, Keerthiram Murugesan, Subhajit Chaudhury +3
Text-based games (TBGs) have emerged as an important collection of NLP tasks, requiring reinforcement learning (RL) agents to combine natural language understanding with reasoning.…
Evidence Aggregation for Answer Re-Ranking in Open-Domain Question Answering
Shuohang Wang, Mo Yu, Jing Jiang +7
A popular recent approach to answering open-domain questions is to first search for question-related passages and then apply reading comprehension models to extract answers. Existi…
Compositional generalization through abstract representations in human and artificial neural networks
Takuya Ito, Tim Klinger, Douglas H. Schultz +3
Humans have a remarkable ability to rapidly generalize to new tasks that is difficult to reproduce in artificial learning systems. Compositionality has been proposed as a key mecha…
Efficient Black-Box Planning Using Macro-Actions with Focused Effects
Cameron Allen, Michael Katz, Tim Klinger +3
The difficulty of deterministic planning increases exponentially with search-tree depth. Black-box planning presents an even greater challenge, since planners must operate without…
e-QRAQ: A Multi-turn Reasoning Dataset and Simulator with Explanations
Clemens Rosenbaum, Tian Gao, Tim Klinger
In this paper we present a new dataset and user simulator e-QRAQ (explainable Query, Reason, and Answer Question) which tests an Agent's ability to read an ambiguous text; ask ques…
Compositional Program Generation for Few-Shot Systematic Generalization
Tim Klinger, Luke Liu, Soham Dan +3
Compositional generalization is a key ability of humans that enables us to learn new concepts from only a handful examples. Neural machine learning models, including the now ubiqui…
Multiresolution Recurrent Neural Networks: An Application to Dialogue Response Generation
Iulian Vlad Serban, Tim Klinger, Gerald Tesauro +4
We introduce the multiresolution recurrent neural network, which extends the sequence-to-sequence framework to model natural language generation as two parallel discrete stochastic…
Learning in Factored Domains with Information-Constrained Visual Representations
Tailia Malloy, Miao Liu, Matthew D. Riemer +3
Humans learn quickly even in tasks that contain complex visual information. This is due in part to the efficient formation of compressed representations of visual information, allo…
Quantifying artificial intelligence through algorithmic generalization
Takuya Ito, Murray Campbell, Lior Horesh +2
The rapid development of artificial intelligence (AI) systems has created an urgent need for their scientific quantification. While their fluency across a variety of domains is imp…
Consolidation via Policy Information Regularization in Deep RL for Multi-Agent Games
Tailia Malloy, Tim Klinger, Miao Liu +3
This paper introduces an information-theoretic constraint on learned policy complexity in the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) reinforcement learning algorit…
A Study of Compositional Generalization in Neural Models
Tim Klinger, Dhaval Adjodah, Vincent Marois +4
Compositional and relational learning is a hallmark of human intelligence, but one which presents challenges for neural models. One difficulty in the development of such models is…
Neural Reasoning Networks: Efficient Interpretable Neural Networks With Automatic Textual Explanations
Stephen Carrow, Kyle Harper Erwin, Olga Vilenskaia +5
Recent advances in machine learning have led to a surge in adoption of neural networks for various tasks, but lack of interpretability remains an issue for many others in which an…
R: Reinforced Reader-Ranker for Open-Domain Question Answering
Shuohang Wang, Mo Yu, Xiaoxiao Guo +7
In recent years researchers have achieved considerable success applying neural network methods to question answering (QA). These approaches have achieved state of the art results i…