Sergey Levine

One-Shot Learning of Manipulation Skills with Online Dynamics Adaptation and Neural Network Priors

2015 · arXiv (Cornell University)

One of the key challenges in applying reinforcement learning to complex robotic control tasks is the need to gather large amounts of experience in order to find an effective policy for the task at hand. …

Imitation from Observation: Learning to Imitate Behaviors from Raw Video via Context Translation

2017 · arXiv (Cornell University)

Imitation learning is an effective approach for autonomous systems to acquire control policies when an explicit reward function is unavailable, using supervision provided as demonstrations from an expert, typically a human operator. However, standard imitation …

MBMF: Model-Based Priors for Model-Free Reinforcement Learning

2017 · arXiv (Cornell University)

Reinforcement Learning is divided in two main paradigms: model-free and model-based. Each of these two paradigms has strengths and limitations, and has been successfully applied to real world domains that are appropriate to its corresponding …

Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods

2018

In this paper, we explore deep reinforcement learning algorithms for vision-based robotic grasping. Model-free deep reinforcement learning (RL) has been successfully applied to a range of challenging environments, but the proliferation of algorithms makes it …

Few-Shot Segmentation Propagation with Guided Networks

2018 · arXiv (Cornell University)

Learning-based methods for visual segmentation have made progress on particular types of segmentation tasks, but are limited by the necessary supervision, the narrow definitions of fixed tasks, and the lack of control during inference for …

Model-Based Reinforcement Learning for Atari

2019 · arXiv (Cornell University)

Model-free reinforcement learning (RL) can be used to learn effective policies for complex tasks, such as Atari games, even from image observations. However, this typically requires very large amounts of interaction -- substantially more, in …

Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model

2019 · arXiv (Cornell University)

Deep reinforcement learning (RL) algorithms can use high-capacity deep networks to learn directly from image observations. However, these high-dimensional observation spaces present a number of challenges in practice, since the policy must now solve two …

Unsupervised Curricula for Visual Meta-Reinforcement Learning

2019 · arXiv (Cornell University)

In principle, meta-reinforcement learning algorithms leverage experience across many tasks to learn fast reinforcement learning (RL) strategies that transfer to similar tasks. However, current meta-RL approaches rely on manually-defined distributions of training tasks, and hand-crafting …

AWAC: Accelerating Online Reinforcement Learning with Offline Datasets

2020 · arXiv (Cornell University)

Reinforcement learning (RL) provides an appealing formalism for learning control policies from experience. However, the classic active formulation of RL necessitates a lengthy active exploration process for each behavior, making it difficult to apply in …

OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learning

2020 · arXiv (Cornell University)

Reinforcement learning (RL) has achieved impressive performance in a variety of online settings in which an agent's ability to query the environment for transitions and rewards is effectively unlimited. However, in many practical applications, the …

One Solution is Not All You Need: Few-Shot Extrapolation via Structured MaxEnt RL

2020 · arXiv (Cornell University)

While reinforcement learning algorithms can learn effective policies for complex tasks, these policies are often brittle to even minor task variations, especially when variations are not explicitly provided during training. One natural approach to this …

Model-Based Visual Planning with Self-Supervised Functional Distances

2020 · arXiv (Cornell University)

A generalist robot must be able to complete a variety of tasks in its environment. One appealing way to specify each task is in terms of a goal observation. However, learning goal-reaching policies with reinforcement …

Evolving Reinforcement Learning Algorithms

2021 · arXiv (Cornell University)

We propose a method for meta-learning reinforcement learning algorithms by searching over the space of computational graphs which compute the loss function for a value-based model-free RL agent to optimize. The learned algorithms are domain-agnostic …

Policy Information Capacity: Information-Theoretic Measure for Task Complexity in Deep Reinforcement Learning

2021 · arXiv (Cornell University)

Progress in deep reinforcement learning (RL) research is largely enabled by benchmark task environments. However, analyzing the nature of those environments is often overlooked. In particular, we still do not have agreeable ways to measure …

Information is Power: Intrinsic Control via Information Capture

2021 · arXiv (Cornell University)

Humans and animals explore their environment and acquire useful skills even in the absence of clear goals, exhibiting intrinsic motivation. The study of intrinsic motivation in artificial agents is concerned with the following question: what …

Context-Aware Language Modeling for Goal-Oriented Dialogue Systems

2022 · arXiv (Cornell University)

Goal-oriented dialogue systems face a trade-off between fluent language generation and task-specific control. While supervised learning with large language models is capable of producing realistic text, how to steer such responses towards completing a specific …

Cal-QL: Calibrated Offline RL Pre-Training for Efficient Online Fine-Tuning

2023 · arXiv (Cornell University)

A compelling use case of offline reinforcement learning (RL) is to obtain a policy initialization from existing datasets followed by fast online fine-tuning with limited interaction. However, existing offline RL methods tend to behave poorly …

Efficient Deep Reinforcement Learning Requires Regulating Overfitting

2023 · arXiv (Cornell University)

Deep reinforcement learning algorithms that learn policies by trial-and-error must learn from limited amounts of data collected by actively interacting with the environment. While many prior works have shown that proper regularization techniques are crucial …

SELFI: Autonomous Self-Improvement with Reinforcement Learning for Social Navigation

2024 · arXiv (Cornell University)

Autonomous self-improving robots that interact and improve with experience are key to the real-world deployment of robotic systems. In this paper, we propose an online learning method, SELFI, that leverages online robot experience to rapidly …

D5RL: Diverse Datasets for Data-Driven Deep Reinforcement Learning

2024 · arXiv (Cornell University)

Offline reinforcement learning algorithms hold the promise of enabling data-driven RL methods that do not require costly or dangerous real-world exploration and benefit from large pre-collected datasets. This in turn can facilitate real-world applications, as …

Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models

2015 · arXiv (Cornell University)

Achieving efficient and scalable exploration in complex domains poses a major challenge in reinforcement learning. While Bayesian and PAC-MDP approaches to the exploration problem offer strong formal guarantees, they are often impractical in higher dimensions …

Automatically Composing Representation Transformations as a Means for Generalization

2019 · International Conference on Learning Representations

A generally intelligent learner should generalize to more complex tasks than it has previously encountered, but the two common paradigms in machine learning -- either training a separate learner per task or training a single …

Do As I Can, Not As I Say: Grounding Language in Robotic Affordances

2022 · arXiv (Cornell University)

Large language models can encode a wealth of semantic knowledge about the world. Such knowledge could be extremely useful to robots aiming to act upon high-level, temporally extended instructions expressed in natural language. However, a …

Inner Monologue: Embodied Reasoning through Planning with Language Models

2022 · arXiv (Cornell University)

Recent works have shown how the reasoning capabilities of Large Language Models (LLMs) can be applied to domains beyond natural language processing, such as planning and interaction for robots. These embodied problems require an agent …