# Learning credit assignment

arXiv: https://arxiv.org/abs/2001.03354 · HTML: https://mlsanity.com/paper/2001.03354/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Jan 2020.

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

- [Statistical mechanics of continual learning: variational principle and mean-field potential](https://mlsanity.com/paper/2212.02846/) (2022)
- [Eight challenges in developing theory of intelligence](https://mlsanity.com/paper/2306.11232/) (2023)
- [Ensemble perspective for understanding temporal credit assignment](https://mlsanity.com/paper/2102.03740/) (2021)
- [Meta predictive learning model of languages in neural circuits](https://mlsanity.com/paper/2309.04106/) (2023)
- [Geometric origin of adversarial vulnerability in deep learning](https://mlsanity.com/paper/2509.01235/) (2025)

## Abstract (arXiv)

Deep learning has achieved impressive prediction accuracies in a variety of scientific and industrial domains. However, the nested non-linear feature of deep learning makes the learning highly non-transparent, i.e., it is still unknown how the learning coordinates a huge number of parameters to achieve a decision making. To explain this hierarchical credit assignment, we propose a mean-field learning model by assuming that an ensemble of sub-networks, rather than a single network, are trained for a classification task. Surprisingly, our model reveals that apart from some deterministic synaptic weights connecting two neurons at neighboring layers, there exist a large number of connections that can be absent, and other connections can allow for a broad distribution of their weight values. Therefore, synaptic connections can be classified into three categories: very important ones, unimportant ones, and those of variability that may partially encode nuisance factors. Therefore, our model learns the credit assignment leading to the decision, and predicts an ensemble of sub-networks that can accomplish the same task, thereby providing insights toward understanding the macroscopic behavior of deep learning through the lens of distinct roles of synaptic weights.