Search Results for author: Kavosh Asadi

Found 17 papers, 7 papers with code

Characterizing the Action-Generalization Gap in Deep Q-Learning

no code implementations11 May 2022 Zhiyuan Zhou, Cameron Allen, Kavosh Asadi, George Konidaris

We study the action generalization ability of deep Q-learning in discrete action spaces.


Coarse-Grained Smoothness for RL in Metric Spaces

no code implementations23 Oct 2021 Omer Gottesman, Kavosh Asadi, Cameron Allen, Sam Lobel, George Konidaris, Michael Littman

We propose a new coarse-grained smoothness definition that generalizes the notion of Lipschitz continuity, is more widely applicable, and allows us to compute significantly tighter bounds on Q-functions, leading to improved learning.

Decision Making

Continuous Doubly Constrained Batch Reinforcement Learning

1 code implementation NeurIPS 2021 Rasool Fakoor, Jonas Mueller, Kavosh Asadi, Pratik Chaudhari, Alexander J. Smola

Reliant on too many experiments to learn good actions, current Reinforcement Learning (RL) algorithms have limited applicability in real-world settings, which can be too expensive to allow exploration.


Learning State Abstractions for Transfer in Continuous Control

2 code implementations8 Feb 2020 Kavosh Asadi, David Abel, Michael L. Littman

In this work, we answer this question in the affirmative, where we take "simple learning algorithm" to be tabular Q-Learning, the "good representations" to be a learned state abstraction, and "challenging problems" to be continuous control tasks.

Continuous Control Q-Learning +1

Lipschitz Lifelong Reinforcement Learning

1 code implementation15 Jan 2020 Erwan Lecarpentier, David Abel, Kavosh Asadi, Yuu Jinnai, Emmanuel Rachelson, Michael L. Littman

We consider the problem of knowledge transfer when an agent is facing a series of Reinforcement Learning (RL) tasks.

reinforcement-learning Transfer Learning

Combating the Compounding-Error Problem with a Multi-step Model

no code implementations30 May 2019 Kavosh Asadi, Dipendra Misra, Seungchan Kim, Michel L. Littman

In this paper, we address the compounding-error problem by introducing a multi-step model that directly outputs the outcome of executing a sequence of actions.

Model-based Reinforcement Learning reinforcement-learning

Mitigating Planner Overfitting in Model-Based Reinforcement Learning

no code implementations3 Dec 2018 Dilip Arumugam, David Abel, Kavosh Asadi, Nakul Gopalan, Christopher Grimm, Jun Ki Lee, Lucas Lehnert, Michael L. Littman

An agent with an inaccurate model of its environment faces a difficult choice: it can ignore the errors in its model and act in the real world in whatever way it determines is optimal with respect to its model.

Model-based Reinforcement Learning reinforcement-learning

Towards a Simple Approach to Multi-step Model-based Reinforcement Learning

no code implementations31 Oct 2018 Kavosh Asadi, Evan Cater, Dipendra Misra, Michael L. Littman

When environmental interaction is expensive, model-based reinforcement learning offers a solution by planning ahead and avoiding costly mistakes.

Model-based Reinforcement Learning reinforcement-learning

Lipschitz Continuity in Model-based Reinforcement Learning

1 code implementation ICML 2018 Kavosh Asadi, Dipendra Misra, Michael L. Littman

We go on to prove an error bound for the value-function estimate arising from Lipschitz models and show that the estimated value function is itself Lipschitz.

Model-based Reinforcement Learning reinforcement-learning

Mean Actor Critic

2 code implementations1 Sep 2017 Cameron Allen, Kavosh Asadi, Melrose Roderick, Abdel-rahman Mohamed, George Konidaris, Michael Littman

We propose a new algorithm, Mean Actor-Critic (MAC), for discrete-action continuous-state reinforcement learning.

Atari Games reinforcement-learning

Hybrid Code Networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning

2 code implementations ACL 2017 Jason D. Williams, Kavosh Asadi, Geoffrey Zweig

End-to-end learning of recurrent neural networks (RNNs) is an attractive solution for dialog systems; however, current techniques are data-intensive and require thousands of dialogs to learn simple behaviors.


Sample-efficient Deep Reinforcement Learning for Dialog Control

no code implementations18 Dec 2016 Kavosh Asadi, Jason D. Williams

Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL).

Policy Gradient Methods reinforcement-learning

An Alternative Softmax Operator for Reinforcement Learning

1 code implementation ICML 2017 Kavosh Asadi, Michael L. Littman

A softmax operator applied to a set of values acts somewhat like the maximization function and somewhat like an average.

Decision Making reinforcement-learning

Cannot find the paper you are looking for? You can Submit a new open access paper.