Michael Amir

Michael Amir

Researcher, Google DeepMind

I'm a researcher at Google DeepMind, where I work on provenance (SynthID). Previously, I was a postdoc at the University of Cambridge, working with the Prorok Lab and affiliated with Trinity College, exploring (multi-)robot learning and watermarking. My PhD explored multi-agent systems on graphs.

Papers

Newest first. * denotes equal contribution. Also on Google Scholar.

CoRL 2026spotlight

Leave It Better Than Found: Value-Guided Cooperation in Sequential Tasks

M. Gomez Andreu, M. Amir, A. Marino, T. An, M. Hutter, A. Prorok

A leader robot rearranges a shared room so that a follower robot can pass
GameSec 2026

Games Over Observation Spaces in Multi-Agent Capture the Flag

M. Frost, M. Amir, S. D. Bopardikar

paper code
Capture the Flag between attackers and defenders on a graph
Preprint 2026

GAMBIT: Learning to Plan Continuous Multi-Robot Trajectories

R. Jain, A. Moldagalieva, L. Magnino, M. Amir, K. Okumura, A. Shankar, W. Hönig, A. Prorok

paper
GAMBIT: continuous multi-robot trajectory planning
Preprint 2026

Remotely Detectable Keyed Communication through Motion

B. Chang, M. Amir, M. Flageat, A. Prorok

paper website
Messages encoded in robot motion and decoded remotely
ICML 2026

Scaling Multi-Agent Environment Co-Design with Diffusion Models

H. X. Li, M. Amir, A. Prorok

paper code
DiCoDe: diffusion-based multi-agent environment co-design
Preprint 2026

Open Problems in Constitutional Preference Reconstruction

E. Clifford*, M. Amir*, A. Findeis, A. Zhao, R. Mullins

paper code
Reconstructing constitutions from preference data
ICLR 2026

Remotely Detectable Robot Policy Watermarking

M. Amir*, M. Flageat*, A. Prorok

paper website
Watermarking pipeline: the owner defines a watermark, the user deploys it on a robot, and an auditor detects it remotely
ICLR 2026

When Is Diversity Rewarded in Cooperative Multi-Agent Learning?

M. Amir*, M. Bettini*, A. Prorok

paper website
Heterogeneous versus homogeneous agent teams and the heterogeneity gain
ICLR 2026oral @ AAAI 2026 WoMAPF

Pairwise is Not Enough: Hypergraph Neural Networks for Multi-Agent Pathfinding

R. Jain, K. Okumura, M. Amir, P. Lio, A. Prorok

paper
Hypergraph neural networks for multi-agent pathfinding
AAAI 2026oral

Graph Attention-Guided Search for Dense Multi-Agent Pathfinding

R. Jain, K. Okumura, M. Amir, A. Prorok

paper code
LaGAT: graph attention-guided search
CoRL 2025

ReCoDe: Reinforcement Learning-based Dynamic Constraint Design for Multi-Agent Coordination

M. Amir, G. Yang, Z. Gao, K. Okumura, H. Woo, A. Prorok

paper video
ReCoDe robot experiment
IEEE T-RO 2025

Time, Travel, and Energy in the Uniform Dispersion Problem

M. Amir, A. M. Bruckstein

paper arxiv
Uniform dispersion of a robot swarm
WAFR 2024

Patrolling Grids with a Bit of Memory

M. Amir*, D. Rabinovich*, A. M. Bruckstein

paper arxiv
A one-bit patrol route on a grid
Theoretical Computer Science 2024

Optimally Reordering Mobile Agents on Parallel Rows

D. Rabinovich*, M. Amir*, A. M. Bruckstein

paper arxiv
Red and blue cars reordered on parallel rows
CDC 2023

Multi-Agent Distributed and Decentralized Geometric Task Allocation

M. Amir, Y. Koifman, Y. Bloch, A. Barel, A. M. Bruckstein

paper arxiv
Agents spreading over a demand density map
IROS 2023

Competitive Ant Coverage: The Value of Pursuit

A. Shats, M. Amir, N. Agmon

paper pdf
A spanning-tree coverage path on a grid
AAMAS 2023

Stigmergy-based, Dual-Layer Coverage of Unknown Regions

O. Rappel*, M. Amir*, A. M. Bruckstein

paper arxiv
Beacon-based coverage of an unknown region
Entropy 2022

A Locust-Inspired Model of Collective Marching on Rings

M. Amir, N. Agmon, A. M. Bruckstein

paper
Locusts marching in a ring-shaped arena
DARS 2021

A Discrete Model of Collective Marching on Rings

M. Amir, N. Agmon, A. M. Bruckstein

paper arxiv
Agents marching on a ring with multiple tracks
RSS 2020

Fast Uniform Dispersion of a Crash-Prone Swarm

M. Amir, A. M. Bruckstein

paper arxiv
A swarm dispersing through a grid environment
Theoretical Computer Science 2019

Probabilistic Pursuits on Graphs

M. Amir, A. M. Bruckstein

paper arxiv
Chain pursuit on a grid graph
AAMAS 2019

Minimizing Travel in the Uniform Dispersal Problem for Robotic Sensors

M. Amir, A. M. Bruckstein

paper arxiv
Robots dispersing through a maze

Contact