A Cordial Sync: Going Beyond Marginal Policies for Multi-Agent Embodied Tasks

By Unnat Jain, Luca Weihs, Eric Kolve, Ali Farhadi, Svetlana Lazebnik, Aniruddha Kembhavi, Alexander Gerhard Schwing. 2020

69 citations recorded by the source corpus.

Autonomous agents must learn to collaborate. It is not scalable to develop a new centralized agent every time a task's difficulty outpaces a single agent's abilities. While multi-agent collaboration research has flourished in gridworld-like environments, relatively little work has considered visually rich domains. Addressing this, we introduce the novel task FurnMove in which agents work together to move a piece of furniture through a living room to a goal. Unlike existing tasks, FurnMove requires agents to coordinate at every timestep. We identify two challenges when training agents to complete FurnMove: existing decentralized action sampling procedures do not permit expressive joint action policies and, in tasks requiring close coordination, the number of failed actions dominates successful actions. To confront these challenges we introduce SYNC-policies (synchronize your actions coherently) and CORDIAL (coordination loss). Using SYNC-policies and CORDIAL, our agents achieve a 58% completion rate on FurnMove, an impressive absolute gain of 25 percentage points over competitive decentralized baselines. Our dataset, code, and pretrained models are available at this https URL .

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

SoundSpaces: Audio-Visual Navigation in 3D Environments (2019, background): “Recent AI agents also aggregate egocentric visual inputs [112,111,63,86,46], often with a spatio-temporal memory [40,79,44,101].”
Building Cooperative Embodied Agents Modularly with Large Language Models (2023, background): “…Samvelyan et al., 2019; Carroll et al., 2019; Puig et al., 2021; 2023), or use uninterpretable continuous vectors (Jiang & Lu, 2018; Das et al., 2019) or limited discrete symbols (Lowe et al., 2017; Jaques et al., 2019; Jain et al., 2020; Patel et al., 2021; Resnick et al., 2018) for communication.”
ProcTHOR: Large-Scale Embodied AI Using Procedural Generation (2022, background): “10 [51] Unnat Jain, Luca Weihs, Eric Kolve, Ali Farhadi, Svetlana Lazebnik, Aniruddha Kembhavi, and Alexander Schwing.”
Visual Room Rearrangement (2021, methodology): “4, follows the same basic structure as is commonly employed within Embod-ied AI tasks [38, 47, 14, 45, 23]: a combination of a convolutional neural network to process input egocentric images, a collection of embedding layers to encode discrete inputs, and an RNN to enable the agent to reason…”
AllenAct: A Framework for Embodied AI Research (2020, methodology): “…[18, 19], language grounding [20, 21], grasping [22, 23], object manipulation [24, 25], future prediction [26], and multi-agent collaboration [27, 28]; as well as using interaction as a tool to learn: environment representations [29], intuitive physics [30], and objects and attributes [31].”
Shaping embodied agent behavior with activity-context priors from egocentric video (2021, background): “…pursuit of such agents has driven exciting new research in visual semantic planning [68], instruction following [58], and object rearrangement [3, 29], typically supported by advanced simulators [45, 32, 57, 22] where policies may be learned quickly and safely before potentially transferring to…”