Overcoming Catastrophic Forgetting by Neuron-level Plasticity Control
To address the issue of catastrophic forgetting in neural networks, we propose a novel, simple, and effective solution called neuron-level plasticity control (NPC). While learning a new task, the proposed method preserves the knowledge for the previous tasks by controlling the plasticity of the network at the neuron level. NPC estimates the importance value of each neuron and consolidates important \textit{neurons} by applying lower learning rates, rather than restricting individual connection weights to stay close to certain values. The experimental results on the incremental MNIST (iMNIST) and incremental CIFAR100 (iCIFAR100) datasets show that neuron-level consolidation is substantially more effective compared to the connection-level consolidation approaches.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Attention-Based Structural-Plasticity
Catastrophic forgetting/interference is a critical problem for lifelong learning machines, which impedes the agents from maintaining their previously learned knowledge while learning new tasks. Neural networks, in partic…
Lifelong learningPermuted-MNISTSplit-MNISTSNAP: Stopping Catastrophic Forgetting in Hebbian Learning with Sigmoidal Neuronal Adaptive Plasticity
Artificial Neural Networks (ANNs) suffer from catastrophic forgetting, where the learning of new tasks causes the catastrophic forgetting of old tasks. Existing Machine Learning (ML) algorithms, including those using Sto…
Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning
Recent studies suggest that Reinforcement Fine-Tuning (RFT) is inherently more resilient to catastrophic forgetting than Supervised Fine-Tuning (SFT). However, whether RFT (e.g., GRPO) can effectively overcome forgetting…
class-incremental learningContinual LearningOn the role of neurogenesis in overcoming catastrophic forgetting
Lifelong learning capabilities are crucial for artificial autonomous agents operating on real-world data, which is typically non-stationary and temporally correlated. In this work, we demonstrate that dynamically grown n…
Incremental LearningLifelong learningSparsity and Heterogeneous Dropout for Continual Learning in the Null Space of Neural Activations
Continual/lifelong learning from a non-stationary input data stream is a cornerstone of intelligence. Despite their phenomenal performance in a wide variety of applications, deep neural networks are prone to forgetting t…
Continual LearningLifelong learning