Skip to main content

Ctrl+K

EIANN Documentation

EIANN Documentation

Using EIANN

Installation
Intro to EIANN
Training EIANN networks
Visualizing EIANN network structure

Galloni et al. 2025 - EIANN models

Feedforward ANN (Backprop)
EI network - Backprop (Dale’s Law, learned inhibition)
EI network - Backprop (Dale’s Law, fixed inhibition)
EI network - Hebbian (learned inh.)
EI network - Dend Target Prop (LDS)
EI network - Dend Target Prop (BTSP)
EI network - BCM

Galloni et al. 2025 - Figures

Figure 1 (diagram): Multi-layer networks with biological specialization
Figure 2: Multilayer learning: Backprop vs Hebb
Figure 3: Dendritic Target Propagation in Rate and Spiking Neurons
Figure 4: Learning Dend-I weights for Dendritic Target Propagation
Figure 5: Receptive Fields
Figure 6: Spirals 2D data
Figure 7: Biological synaptic updates
Figure 8: Learning top-down weights

Galloni et al. 2025 - Supplementary Figures

Figure S1: EI dynamics
Figure S2: Soma-I representations
Figure S3 - H2E confusion
Figure S4: H2E receptive fields
Figure S5: Robustness to perturbations
Figure S6: Neural representations with biological plasticity rules
Figure S7: Receptive fields with biological plasticity
Supplementary Tables: Accuracy
Supplementary Tables: Model Hyperparameters

Python API

YAML Config Glossary
EIANN

Colab

Repository
Suggest edit
Open issue

.ipynb

Figure 1 (diagram): Multi-layer networks with biological specialization

Figure 1 (diagram): Multi-layer networks with biological specialization#

Diagram of EIANN network architecture

previous

EI network - BCM

next

Figure 2: Multilayer learning: Backprop vs Hebb

By Alessandro Galloni and Aaron Milstein

© Copyright 2025.