split-mnist

Here are 3 public repositories matching this topic...

GMvandeVen / brain-inspired-replay

A brain-inspired version of generative replay for continual learning with deep neural networks (e.g., class-incremental learning on CIFAR-100; PyTorch code).

deep-learning artificial-neural-networks replay incremental-learning variational-autoencoder lifelong-learning distillation brain-inspired continual-learning elastic-weight-consolidation replay-through-feedback split-mnist generative-replay permuted-mnist split-cifar100 internal-replay synaptic-intelligence

Updated Jul 6, 2023
Python

GongJae00 / Router-Free-Developmental-Specialization

Star

Router-free neural specialization with capacity-selection-composition diagnostics for emergent expert routing under strict activation-derived inference.

pytorch continual-learning mixture-of-experts split-mnist mechanistic-interpretability rfds router-free emergent-routing neural-specialization capacity-selection-composition activation-derived-routing split-cifar10

Updated Jun 27, 2026
Python

Ayai1 / continual-learning-dce-cvae

Star

Class-incremental continual learning on Split-MNIST using pure latent replay - a DCE-CVAE backbone with a four-tier memory hierarchy and per-task orthogonal latent subspaces. Replays compressed latents instead of raw images. ~0.66 average accuracy over 5 sequential tasks vs ~0.20 for naive fine-tuning.

deep-learning pytorch cvae variational-autoencoder lifelong-learning continual-learning catastrophic-forgetting split-mnist latent-replay

Updated Sep 7, 2026
Python

Add this topic to your repo

To associate your repository with the split-mnist topic, visit your repo's landing page and select "manage topics."

Learn more