# Predicting Multi-Agent Specialization via Task Parallelizability

arXiv: https://arxiv.org/abs/2503.15703 · HTML: https://mlsanity.com/paper/2503.15703/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 67 references that are in the map.
- First posted on arXiv 19 Mar 2025.

## Builds on (most cited first, 24 of 24)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments](https://mlsanity.com/paper/1706.02275/) (2017)
- [On the Utility of Learning about Humans for Human-AI Coordination](https://mlsanity.com/paper/1910.05789/) (2019)
- [Is Independent Learning All You Need in the StarCraft Multi-Agent Challenge?](https://mlsanity.com/paper/2011.09533/) (2020)
- [A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration](https://mlsanity.com/paper/2310.02170/) (2023)
- [Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration](https://mlsanity.com/paper/2307.05300/) (2023)
- [ROMA: Multi-Agent Reinforcement Learning with Emergent Roles](https://mlsanity.com/paper/2003.08039/) (2020)
- [RODE: Learning Roles to Decompose Multi-Agent Tasks](https://mlsanity.com/paper/2010.01523/) (2020)
- [Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffolding](https://mlsanity.com/paper/2401.12954/) (2024)
- [Discovered Policy Optimisation](https://mlsanity.com/paper/2210.05639/) (2022)
- [Celebrating Diversity in Shared Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2106.02195/) (2021)
- [A multi-agent reinforcement learning model of common-pool resource appropriation](https://mlsanity.com/paper/1707.06600/) (2017)
- [On the Importance of Environments in Human-Robot Coordination](https://mlsanity.com/paper/2106.10853/) (2021)
- [Heterogeneous Multi-Robot Reinforcement Learning](https://mlsanity.com/paper/2301.07137/) (2023)
- [Feudal Multi-Agent Hierarchies for Cooperative Reinforcement Learning](https://mlsanity.com/paper/1901.08492/) (2019)
- [Understanding Human Intelligence through Human Limitations](https://mlsanity.com/paper/2009.14050/) (2020)
- [Quantifying the effects of environment and population diversity in multi-agent reinforcement learning](https://mlsanity.com/paper/2102.08370/) (2021)
- [Controlling Behavioral Diversity in Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2405.15054/) (2024)
- [The Overcooked Generalisation Challenge: Evaluating Cooperation with Novel Partners in Unknown Environments Using Unsupervised Environment Design](https://mlsanity.com/paper/2406.17949/) (2024)
- [Capability-aware Task Allocation and Team Formation Analysis for Cooperative Exploration of Complex Environments](https://mlsanity.com/paper/2411.00400/) (2024)
- [Breaking the mold: The challenge of large scale MARL specialization](https://mlsanity.com/paper/2410.02128/) (2024)

## Built on by (most cited first, 5 of 5)

- [Latent Collaboration in Multi-Agent Systems](https://mlsanity.com/paper/2511.20639/) (2025)
- [Inherent and emergent liability issues in LLM-based agentic systems: a principal-agent perspective](https://mlsanity.com/paper/2504.03255/) (2025)
- [MASPO: Joint Prompt Optimization for LLM-based Multi-Agent Systems](https://mlsanity.com/paper/2605.06623/) (2026)
- [StateBridge: Training-free Hidden-state Alignment for Latent Communication in LLM Multi-Agent Systems](https://mlsanity.com/paper/2608.13317/) (2026)
- [JaxAHT: A JAX-Based Library for Ad Hoc Teamwork](https://mlsanity.com/paper/2609.13716/) (2026)

## Abstract (arXiv)

When should we encourage specialization in multi-agent systems versus train generalists that perform the entire task independently? We propose that specialization largely depends on task parallelizability: the potential for multiple agents to execute task components concurrently. Drawing inspiration from Amdahl's Law in distributed systems, we present a closed-form bound that predicts when specialization improves performance, depending only on task concurrency and team size. We validate our model on two standard MARL benchmarks that represent opposite regimes -- StarCraft Multi-Agent Challenge (SMAC, unlimited concurrency) and Multi-Particle Environment (MPE, unit-capacity bottlenecks) -- and observe close alignment between the bound at each extreme and an empirical measure of specialization. Three follow-up experiments in Overcooked-AI demonstrate that the model works in environments with more complex spatial and resource bottlenecks that allow for a range of strategies. Beyond prediction, the bound also serves as a diagnostic tool, highlighting biases in MARL training algorithms that cause sub-optimal convergence to specialist strategies with larger state spaces.