307 citations · 724 across the 15 of their papers we have counts for
20 papers
Composer 2 Technical Report
Cursor Research, :, Aaron Chan +53
Composer 2 is a specialized model designed for agentic software engineering. The model demonstrates strong long-term planning and coding intelligence while maintaining the ability…
GPT-4 Technical Report
OpenAI, Josh Achiam, Steven Adler +278
We report the development of GPT-4, a large-scale, multimodal model which can accept image and text inputs and produce text outputs. While less capable than humans in many real-wor…
Learning on the Job: Self-Rewarding Offline-to-Online Finetuning for Industrial Insertion of Novel Connectors from Vision
Ashvin Nair, Brian Zhu, Gokul Narayanan +2
Learning-based methods in robotics hold the promise of generalization, but what can be done if a learned policy does not generalize to a new situation? In principle, if an agent ca…
Generalization with Lossy Affordances: Leveraging Broad Offline Data for Learning Visuomotor Tasks
Kuan Fang, Patrick Yin, Ashvin Nair +3
The utilization of broad datasets has proven to be crucial for generalization for a wide range of fields. However, how to effectively make use of diverse multi-task data for novel…
Bisimulation Makes Analogies in Goal-Conditioned Reinforcement Learning
Philippe Hansen-Estruch, Amy Zhang, Ashvin Nair +2
Building generalizable goal-conditioned agents from rich observations is a key to reinforcement learning (RL) solving real world problems. Traditionally in goal-conditioned RL, an…
Planning to Practice: Efficient Online Fine-Tuning by Composing Goals in Latent Space
Kuan Fang, Patrick Yin, Ashvin Nair +1
General-purpose robots require diverse repertoires of behaviors to complete challenging tasks in real-world unstructured environments. To address this issue, goal-conditioned reinf…
Offline Reinforcement Learning with Implicit Q-Learning
Ilya Kostrikov, Ashvin Nair, Sergey Levine
Offline reinforcement learning requires reconciling two conflicting aims: learning a policy that improves over the behavior policy that collected the dataset, while at the same tim…
Offline Meta-Reinforcement Learning with Online Self-Supervision
Vitchyr H. Pong, Ashvin Nair, Laura Smith +2
Meta-reinforcement learning (RL) methods can meta-train policies that adapt to new tasks with orders of magnitude less data than standard RL, but meta-training itself is costly and…
What Can I Do Here? Learning New Skills by Imagining Visual Affordances
Alexander Khazatsky, Ashvin Nair, Daniel Jing +1
A generalist robot equipped with learned skills must be able to perform many tasks in many different environments. However, zero-shot generalization to new settings is not always p…
DisCo RL: Distribution-Conditioned Reinforcement Learning for General-Purpose Policies
Soroush Nasiriany, Vitchyr H. Pong, Ashvin Nair +3
Can we use reinforcement learning to learn general-purpose policies that can perform a wide range of different tasks, resulting in flexible and reusable skills? Contextual policies…
AWAC: Accelerating Online Reinforcement Learning with Offline Datasets
Ashvin Nair, Abhishek Gupta, Murtaza Dalal +1
Reinforcement learning (RL) provides an appealing formalism for learning control policies from experience. However, the classic active formulation of RL necessitates a lengthy acti…
Meta-Reinforcement Learning for Robotic Industrial Insertion Tasks
Gerrit Schoettler, Ashvin Nair, Juan Aparicio Ojea +2
Robotic insertion tasks are characterized by contact and friction mechanics, making them challenging for conventional feedback control methods due to unmodeled physical effects. Re…
Contextual Imagined Goals for Self-Supervised Robotic Learning
Ashvin Nair, Shikhar Bahl, Alexander Khazatsky +3
While reinforcement learning provides an appealing formalism for learning individual skills, a general-purpose robotic system must be able to master an extensive repertoire of beha…
Deep Reinforcement Learning for Industrial Insertion Tasks with Visual Inputs and Natural Rewards
Gerrit Schoettler, Ashvin Nair, Jianlan Luo +4
Connector insertion and many other tasks commonly found in modern manufacturing settings involve complex contact dynamics and friction. Since it is difficult to capture related phy…
Skew-Fit: State-Covering Self-Supervised Reinforcement Learning
Vitchyr H. Pong, Murtaza Dalal, Steven Lin +3
Autonomous agents that must exhibit flexible and broad capabilities will need to be equipped with large repertoires of skills. Defining each skill with a manually-designed reward f…
Residual Reinforcement Learning for Robot Control
Tobias Johannink, Shikhar Bahl, Ashvin Nair +6
Conventional feedback control methods can solve various types of robot control problems very efficiently by capturing the structure with explicit models, such as rigid body equatio…
Visual Reinforcement Learning with Imagined Goals
Ashvin Nair, Vitchyr Pong, Murtaza Dalal +3
For an autonomous agent to fulfill a wide range of user-specified goals at test time, it must be able to learn broadly applicable and general-purpose skill repertoires. Furthermore…
Overcoming Exploration in Reinforcement Learning with Demonstrations
Ashvin Nair, Bob McGrew, Marcin Andrychowicz +2
Exploration in environments with sparse rewards has been a persistent problem in reinforcement learning (RL). Many tasks are natural to specify with a sparse reward, and manually s…
Combining Self-Supervised Learning and Imitation for Vision-Based Rope Manipulation
Ashvin Nair, Dian Chen, Pulkit Agrawal +4
Manipulation of deformable objects, such as ropes and cloth, is an important but challenging problem in robotics. We present a learning-based system where a robot takes as input a…
Learning to Poke by Poking: Experiential Learning of Intuitive Physics
Pulkit Agrawal, Ashvin Nair, Pieter Abbeel +2
We investigate an experiential learning paradigm for acquiring an internal model of intuitive physics. Our model is evaluated on a real-world robotic manipulation task that require…