PathNet: Evolution Channels Gradient Descent in Super Neural Networks
read the original abstract
For artificial general intelligence (AGI) it would be efficient if multiple users trained the same giant neural network, permitting parameter reuse, without catastrophic forgetting. PathNet is a first step in this direction. It is a neural network algorithm that uses agents embedded in the neural network whose task is to discover which parts of the network to re-use for new tasks. Agents are pathways (views) through the network which determine the subset of parameters that are used and updated by the forwards and backwards passes of the backpropogation algorithm. During learning, a tournament selection genetic algorithm is used to select pathways through the neural network for replication and mutation. Pathway fitness is the performance of that pathway measured according to a cost function. We demonstrate successful transfer learning; fixing the parameters along a path learned on task A and re-evolving a new population of paths for task B, allows task B to be learned faster than it could be learned from scratch or after fine-tuning. Paths evolved on task B re-use parts of the optimal path evolved on task A. Positive transfer was demonstrated for binary MNIST, CIFAR, and SVHN supervised learning classification tasks, and a set of Atari and Labyrinth reinforcement learning tasks, suggesting PathNets have general applicability for neural network training. Finally, PathNet also significantly improves the robustness to hyperparameter choices of a parallel asynchronous reinforcement learning algorithm (A3C).
This paper has not been read by Pith yet.
Forward citations
Cited by 14 Pith papers
-
MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data
MoRe identifies modular structure in representations themselves to enable principled reuse, alignment, and expansion of modules during continual adaptation on sequential data.
-
MILE: Mixture of Incremental LoRA Experts for Continual Semantic Segmentation across Domains and Modalities
MILE combines incremental LoRA experts with prototype-guided gating to support continual semantic segmentation across domains and modalities while adding only a small number of parameters per task.
-
Learning Without Losing Identity: Capability Evolution for Embodied Agents
Embodied agents maintain a persistent identity while evolving capabilities via modular ECMs, raising simulated task success from 32.4% to 91.3% over 20 iterations with zero policy drift or safety violations.
-
Learning Without Losing Identity: Capability Evolution for Embodied Agents
Embodied agents maintain persistent identity while evolving modular capabilities through a closed-loop process, raising simulated task success from 32.4% to 91.3% with zero policy drift.
-
Evidence of an Emergent "Self" in Continual Robot Learning
Continual learning robots form a significantly more stable invariant subnetwork than constant-task controls, and preserving it improves adaptation while damaging it hurts performance.
-
TACO: Temporal Consensus Optimization for Continual Neural Mapping
TACO reformulates neural implicit mapping as temporal consensus optimization to enable continual adaptation to scene changes without data replay or storage.
-
A Survey of Continual Reinforcement Learning
The paper surveys CRL literature, proposes a taxonomy of methods into four categories based on knowledge storage and transfer, reviews metrics and benchmarks, and outlines challenges and future research directions.
-
Continual Reinforcement Learning with Diversity Exploration and Adversarial Self-Correction
CDAN framework uses diversity exploration and adversarial self-correction for continual RL in continuous control, evaluated on new CAM environment with NSD metric showing 18.35% NSD improvement over baseline.
-
On the Stability of Growth in Structural Plasticity
Newborn units in growing neural networks are forward-active but backward-starved, receiving weaker gradients than existing units and creating integration challenges that make growth less reliable than pruning in compl...
-
MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data
MoRe decomposes representations into identifiable hierarchical modules to enable principled continual adaptation on sequential data.
-
MoRe: Modular Representations for Principled Continual Representation Learning on Sequential Data
MoRe identifies modular representations in sequential data for continual learning with identifiability guarantees, enabling principled adaptation without disrupting old modules.
-
Incremental learning for audio classification with Hebbian Deep Neural Networks
A kernel plasticity approach in Hebbian DNNs for incremental sound classification achieves 76.3% accuracy over five steps on ESC-50, outperforming the 68.7% baseline without plasticity.
-
ARROW: Augmented Replay for RObust World models
ARROW adds a distribution-matching long-term replay buffer to DreamerV3 and shows reduced forgetting versus same-size baselines on Atari and Procgen continual RL benchmarks.
-
Adaptive Reorganization of Neural Pathways for Continual Learning with Spiking Neural Networks
SOR-SNN employs Self-Organizing Regulation networks to reorganize a single SNN into sparse pathways, achieving better performance, energy efficiency, memory use, backward transfer, and self-repair on continual learnin...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.