Zobrazeno 1 - 10
of 338
pro vyhledávání: '"P. Sodhani"'
In the age of artificial intelligence, the role of large language models (LLMs) is becoming increasingly central. Despite their growing prevalence, their capacity to consolidate knowledge from different training documents - a crucial ability in numer
Externí odkaz:
http://arxiv.org/abs/2310.15372
Autor:
Klissarov, Martin, D'Oro, Pierluca, Sodhani, Shagun, Raileanu, Roberta, Bacon, Pierre-Luc, Vincent, Pascal, Zhang, Amy, Henaff, Mikael
Exploring rich environments and evaluating one's actions without prior knowledge is immensely challenging. In this paper, we propose Motif, a general method to interface such prior knowledge from a Large Language Model (LLM) with an agent. Motif is b
Externí odkaz:
http://arxiv.org/abs/2310.00166
Autor:
Bou, Albert, Bettini, Matteo, Dittert, Sebastian, Kumar, Vikash, Sodhani, Shagun, Yang, Xiaomeng, De Fabritiis, Gianni, Moens, Vincent
Striking a balance between integration and modularity is crucial for a machine learning library to be versatile and user-friendly, especially in handling decision and control tasks that involve large development teams and complex, real-world data, an
Externí odkaz:
http://arxiv.org/abs/2306.00577
Offline reinforcement learning (RL) allows agents to learn effective, return-maximizing policies from a static dataset. Three popular algorithms for offline RL are Conservative Q-Learning (CQL), Behavior Cloning (BC), and Decision Transformer (DT), f
Externí odkaz:
http://arxiv.org/abs/2305.14550
Autor:
Ma, Yecheng Jason, Sodhani, Shagun, Jayaraman, Dinesh, Bastani, Osbert, Kumar, Vikash, Zhang, Amy
Reward and representation learning are two long-standing challenges for learning an expanding set of robot manipulation skills from sensory observations. Given the inherent cost and scarcity of in-domain, task-specific robot data, learning from large
Externí odkaz:
http://arxiv.org/abs/2210.00030
Our theoretical understanding of deep learning has not kept pace with its empirical success. While network architecture is known to be critical, we do not yet understand its effect on learned representations and network behavior, or how this architec
Externí odkaz:
http://arxiv.org/abs/2207.10430
Autor:
Sodhani, Shagun, Faramarzi, Mojtaba, Mehta, Sanket Vaibhav, Malviya, Pranshu, Abdelsalam, Mohamed, Janarthanan, Janarthanan, Chandar, Sarath
This primer is an attempt to provide a detailed summary of the different facets of lifelong learning. We start with Chapter 2 which provides a high-level overview of lifelong learning systems. In this chapter, we discuss prominent scenarios in lifelo
Externí odkaz:
http://arxiv.org/abs/2207.04354
Reinforcement learning (RL) agents need to be robust to variations in safety-critical environments. While system identification methods provide a way to infer the variation from online experience, they can fail in settings where fast identification i
Externí odkaz:
http://arxiv.org/abs/2202.07013
In reinforcement learning (RL), when defining a Markov Decision Process (MDP), the environment dynamics is implicitly assumed to be stationary. This assumption of stationarity, while simplifying, can be unrealistic in many scenarios. In the continual
Externí odkaz:
http://arxiv.org/abs/2110.06972
The benefit of multi-task learning over single-task learning relies on the ability to use relations across tasks to improve performance on any single task. While sharing representations is an important mechanism to share information across tasks, its
Externí odkaz:
http://arxiv.org/abs/2102.06177