# An Economy of Neural Networks: Learning from Heterogeneous Experiences

arXiv: https://arxiv.org/abs/2110.11582 · HTML: https://mlsanity.com/paper/2110.11582/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Oct 2021.

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

- [Simulating the Economic Impact of Rationality through Reinforcement Learning and Agent-Based Modelling](https://mlsanity.com/paper/2405.02161/) (2024)
- [EconGym: A Scalable AI Testbed with Diverse Economic Tasks](https://mlsanity.com/paper/2506.12110/) (2025)
- [Think, Speak, Decide: Language-Augmented Multi-Agent Reinforcement Learning for Economic Decision-Making](https://mlsanity.com/paper/2511.12876/) (2025)
- [Learning Macroeconomic Policies through Dynamic Stackelberg Mean-Field Games](https://mlsanity.com/paper/2403.12093/) (2024)

## Abstract (arXiv)

This paper proposes a new way to model behavioral agents in dynamic macro-financial environments. Agents are described as neural networks and learn policies from idiosyncratic past experiences. I investigate the feedback between irrationality and past outcomes in an economy with heterogeneous shocks similar to Aiyagari (1994). In the model, the rational expectations assumption is seriously violated because learning of a decision rule for savings is unstable. Agents who fall into learning traps save either excessively or save nothing, which provides a candidate explanation for several empirical puzzles about wealth distribution. Neural network agents have a higher average MPC and exhibit excess sensitivity of consumption. Learning can negatively affect intergenerational mobility.