# Diffusing Blame: Task-Dependent Credit Assignment in Biologically Plausible Dual-Stream Networks

arXiv: https://arxiv.org/abs/2606.31700 · HTML: https://mlsanity.com/paper/2606.31700/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods.
- Builds on 16 of its 35 references that are in the map.
- First posted on arXiv 30 Jun 2026.

## Placed in

- Deep reinforcement learning methods / [Deep RL algorithms and representations](https://mlsanity.com/deeprl/deep-rl-algorithms-and-representations/)

## Builds on (most cited first, 16 of 16)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Evolution Strategies as a Scalable Alternative to Reinforcement Learning](https://mlsanity.com/paper/1703.03864/) (2017)
- [Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulation](https://mlsanity.com/paper/2106.13281/) (2021)
- [Direct Feedback Alignment Provides Learning in Deep Neural Networks](https://mlsanity.com/paper/1609.01596/) (2016)
- [Difference Target Propagation](https://mlsanity.com/paper/1412.7525/) (2014)
- [Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures](https://mlsanity.com/paper/1807.04587/) (2018)
- [Dendritic cortical microcircuits approximate the backpropagation algorithm](https://mlsanity.com/paper/1810.11393/) (2018)
- [Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement Learning](https://mlsanity.com/paper/2402.16801/) (2024)
- [Towards Biologically Plausible Deep Learning](https://mlsanity.com/paper/1502.04156/) (2015)
- [How Important is Weight Symmetry in Backpropagation?](https://mlsanity.com/paper/1510.05067/) (2015)
- [Direct Feedback Alignment Scales to Modern Deep Learning Tasks and Architectures](https://mlsanity.com/paper/2006.12878/) (2020)
- [Principled Training of Neural Networks with Direct Feedback Alignment](https://mlsanity.com/paper/1906.04554/) (2019)
- [Review and Comparison of Commonly Used Activation Functions for Deep Neural Networks](https://mlsanity.com/paper/2010.09458/) (2020)
- [Silicon Photonic Architecture for Training Deep Neural Networks with Direct Feedback Alignment](https://mlsanity.com/paper/2111.06862/) (2021)
- [Deep Learning Activation Functions: Fixed-Shape, Parametric, Adaptive, Stochastic, Miscellaneous, Non-Standard, Ensemble](https://mlsanity.com/paper/2407.11090/) (2024)
- [A Diagnostic Evaluation of Neural Networks Trained with the Error Diffusion Learning Algorithm](https://mlsanity.com/paper/2504.14814/) (2025)

## Built on by (most cited first, 1 of 1)

- [Common-Mode Collapse and Recovery in Direct Feedback Alignment](https://mlsanity.com/paper/2609.31589/) (2026)

## Abstract (arXiv)

Biological neural circuits obey Dale's principle: each neuron's synapses are uniformly excitatory or inhibitory. Artificial networks that respect this constraint must coordinate separate excitatory and inhibitory populations, fundamentally changing how credit is assigned during learning. Several biologically plausible learning rules avoid backpropagation's weight transport requirement, but it has been difficult to achieve strong performance under Dale's principle beyond MNIST. Error Diffusion (ED) was originally proposed in a dual-stream excitatory/inhibitory architecture, where learning is driven by routing global error signals to all layers without transporting transposed forward weights or relying on random feedback matrices. Whether such a rule can scale under Dale's principle across both supervised classification and reinforcement learning remains unknown. Here, we introduce modulo error routing to extend Error Diffusion beyond binary classification, and show that a dual-stream excitatory/inhibitory architecture trained with this method achieves 96.7% on MNIST and establishes a 61.7% baseline on CIFAR-10, demonstrating that representation learning is possible even when strictly enforcing Dale's principle. For the classification setting, we introduce three domain-specific innovations: layer-specific sigmoid widths, batch-centered class error signals, and asymmetric initialization, and ablation analysis reveals that their relative importance reverses between MNIST and CIFAR-10, exposing task-dependent credit-assignment bottlenecks invisible to single-benchmark evaluation. In reinforcement learning, we integrate ED with Proximal Policy Optimization (PPO) and evaluate it on continuous-control tasks in Google Brax and on Craftax, an open-ended exploration task. We show that ED-PPO achieves competitive performance relative to Direct Feedback Alignment, a backpropagation-free baseline.