DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning

• URL: http://arxiv.org/abs/2406.11896v1
• Date: Fri, 14 Jun 2024 17:49:55 GMT
• Title: DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning
• Authors: Hao Bai, Yifei Zhou, Mert Cemri, Jiayi Pan, Alane Suhr, Sergey Levine, Aviral Kumar,
• Abstract summary: This paper introduces a novel autonomous RL approach, called DigiRL, for training in-the-wild device control agents. We build a scalable and parallelizable Android learning environment equipped with a VLM-based evaluator. We demonstrate the effectiveness of DigiRL using the Android-in-the-Wild dataset, where our 1.3B VLM trained with RL achieves a 49.5% absolute improvement.
• Score: 61.10299147201369
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Training corpuses for vision language models (VLMs) typically lack sufficient amounts of decision-centric data. This renders off-the-shelf VLMs sub-optimal for decision-making tasks such as in-the-wild device control through graphical user interfaces (GUIs). While training with static demonstrations has shown some promise, we show that such methods fall short for controlling real GUIs due to their failure to deal with real-world stochasticity and non-stationarity not captured in static observational data. This paper introduces a novel autonomous RL approach, called DigiRL, for training in-the-wild device control agents through fine-tuning a pre-trained VLM in two stages: offline RL to initialize the model, followed by offline-to-online RL. To do this, we build a scalable and parallelizable Android learning environment equipped with a VLM-based evaluator and develop a simple yet effective RL approach for learning in this domain. Our approach runs advantage-weighted RL with advantage estimators enhanced to account for stochasticity along with an automatic curriculum for deriving maximal learning signal. We demonstrate the effectiveness of DigiRL using the Android-in-the-Wild (AitW) dataset, where our 1.3B VLM trained with RL achieves a 49.5% absolute improvement -- from 17.7 to 67.2% success rate -- over supervised fine-tuning with static human demonstration data. These results significantly surpass not only the prior best agents, including AppAgent with GPT-4V (8.3% success rate) and the 17B CogAgent trained with AitW data (38.5%), but also the prior best autonomous RL approach based on filtered behavior cloning (57.8%), thereby establishing a new state-of-the-art for digital agents for in-the-wild device control.

Related papers

• DistRL: An Asynchronous Distributed Reinforcement Learning Framework for On-Device Control Agents [38.0441002097771]
  DistRL is a novel framework designed to enhance the efficiency of online RL fine-tuning for mobile device control agents. On average, DistRL delivers a 3X improvement in training efficiency and enables training data collection 2.4X faster than the leading synchronous multi-machine methods.
  arXiv  Detail & Related papers  (2024-10-18T18:19:56Z)
• Scaling Offline Model-Based RL via Jointly-Optimized World-Action Model Pretraining [49.730897226510095]
  We introduce JOWA: Jointly-Reinforced World-Action model, an offline model-based RL agent pretrained on Atari games with 6 billion tokens data. Our largest agent, with 150 million parameters, 78.9% human-level performance on pretrained games using only 10% subsampled offline data, outperforming existing state-of-the-art large-scale offline RL baselines by 31.6% on averange.
  arXiv  Detail & Related papers  (2024-10-01T10:25:03Z)
• An Extremely Data-efficient and Generative LLM-based Reinforcement Learning Agent for Recommenders [1.0154385852423122]
  reinforcement learning (RL) algorithms have been instrumental in maximizing long-term customer satisfaction and avoiding short-term, myopic goals in industrial recommender systems. The goal is to train an RL agent to maximize the purchase reward given a detailed human instruction describing a desired product. This report also evaluates the RL agents trained using generative trajectories.
  arXiv  Detail & Related papers  (2024-08-28T10:31:50Z)
• SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning [85.21378553454672]
  We develop a library containing a sample efficient off-policy deep RL method, together with methods for computing rewards and resetting the environment. We find that our implementation can achieve very efficient learning, acquiring policies for PCB board assembly, cable routing, and object relocation. These policies achieve perfect or near-perfect success rates, extreme robustness even under perturbations, and exhibit emergent robustness recovery and correction behaviors.
  arXiv  Detail & Related papers  (2024-01-29T10:01:10Z)
• A Real-World Quadrupedal Locomotion Benchmark for Offline Reinforcement Learning [27.00483962026472]
  We benchmark 11 offline reinforcement learning algorithms in realistic quadrupedal locomotion dataset. Experiments show that the best-performing ORL algorithms can achieve competitive performance compared with the model-free RL. Our proposed benchmark will serve as a development platform for testing and evaluating the performance of ORL algorithms in real-world legged locomotion tasks.
  arXiv  Detail & Related papers  (2023-09-13T13:18:29Z)
• Real-Time Model-Free Deep Reinforcement Learning for Force Control of a Series Elastic Actuator [56.11574814802912]
  State-of-the art robotic applications utilize series elastic actuators (SEAs) with closed-loop force control to achieve complex tasks such as walking, lifting, and manipulation. Model-free PID control methods are more prone to instability due to nonlinearities in the SEA. Deep reinforcement learning has proved to be an effective model-free method for continuous control tasks.
  arXiv  Detail & Related papers  (2023-04-11T00:51:47Z)
• Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels [112.63440666617494]
  Reinforcement learning algorithms can succeed but require large amounts of interactions between the agent and the environment. We propose a new method to solve it, using unsupervised model-based RL, for pre-training the agent. We show robust performance on the Real-Word RL benchmark, hinting at resiliency to environment perturbations during adaptation.
  arXiv  Detail & Related papers  (2022-09-24T14:22:29Z)
• AWAC: Accelerating Online Reinforcement Learning with Offline Datasets [84.94748183816547]
  We show that our method, advantage weighted actor critic (AWAC), enables rapid learning of skills with a combination of prior demonstration data and online experience. Our results show that incorporating prior data can reduce the time required to learn a range of robotic skills to practical time-scales.
  arXiv  Detail & Related papers  (2020-06-16T17:54:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.