# Distributionally Robust Free Energy Principle for Decision-Making

arXiv: https://arxiv.org/abs/2503.13223 · HTML: https://mlsanity.com/paper/2503.13223/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 63 references that are in the map.
- First posted on arXiv 17 Mar 2025.

## Builds on (most cited first, 15 of 15)

- [Maximum Entropy RL (Provably) Solves Some Robust RL Problems](https://mlsanity.com/paper/2103.06257/) (2021)
- [Building Machines that Learn and Think with People](https://mlsanity.com/paper/2408.03943/) (2024)
- [Variational Neural Annealing](https://mlsanity.com/paper/2101.10154/) (2021)
- [Towards quantum enhanced adversarial robustness in machine learning](https://mlsanity.com/paper/2306.12688/) (2023)
- [The free energy principle made simpler but not too simple](https://mlsanity.com/paper/2201.06387/) (2022)
- [Contrastive Active Inference](https://mlsanity.com/paper/2110.10083/) (2021)
- [Towards a Theory of Evolution as Multilevel Learning](https://mlsanity.com/paper/2110.14602/) (2021)
- [Collective behavior from surprise minimization](https://mlsanity.com/paper/2307.14804/) (2023)
- [Maximum diffusion reinforcement learning](https://mlsanity.com/paper/2309.15293/) (2023)
- [Distributionally Robust Linear Quadratic Control](https://mlsanity.com/paper/2305.17037/) (2023)
- [The Two Kinds of Free Energy and the Bayesian Revolution](https://mlsanity.com/paper/2004.11763/) (2020)
- [Probabilistic design of optimal sequential decision-making algorithms in learning and control](https://mlsanity.com/paper/2201.05212/) (2022)
- [On the Statistical Consistency of Risk-Sensitive Bayesian Decision-Making](https://mlsanity.com/paper/1903.05220/) (2019)
- [Distributionally Robust Model Predictive Control: Closed-loop Guarantees and Scalable Algorithms](https://mlsanity.com/paper/2309.12758/) (2023)
- [On Convex Data-Driven Inverse Optimal Control for Nonlinear, Non-stationary and Stochastic Systems](https://mlsanity.com/paper/2306.13928/) (2023)

## Built on by (most cited first, 4 of 4)

- [An Energy-Based Mechanism for Compositional Behavior](https://mlsanity.com/paper/2512.04745/) (2025)
- [Energy-Based Dynamical Models for Neurocomputation, Learning, and Optimization](https://mlsanity.com/paper/2604.05042/) (2026)
- [A Game Theoretic Free Energy Analysis of Higher Order Synergy in Attention Heads of Large Language Models](https://mlsanity.com/paper/2605.09515/) (2026)
- [Agent-Centric Social Trajectory Prediction: A Free Energy Principle Perspective](https://mlsanity.com/paper/2605.25748/) (2026)

## Abstract (arXiv)

Despite their groundbreaking performance, autonomous agents can misbehave when training and environmental conditions become inconsistent, with minor mismatches leading to undesirable behaviors or even catastrophic failures. Robustness towards these training-environment ambiguities is a core requirement for intelligent agents and its fulfillment is a long-standing challenge towards their real-world deployments. Here, we introduce a Distributionally Robust Free Energy model (DR-FREE) that instills this core property by design. Combining a robust extension of the free energy principle with a resolution engine, DR-FREE wires robustness into the agent decision-making mechanisms. Across benchmark experiments, DR-FREE enables the agents to complete the task even when, in contrast, state-of-the-art models fail. This milestone may inspire both deployments in multi-agent settings and, at a perhaps deeper level, the quest for an explanation of how natural agents -- with little or no training -- survive in capricious environments.