Looking through the past: better knowledge retention for generative replay in continual learning
In this work, we improve the generative replay in a continual learning setting to perform well on challenging scenarios. Current generative rehearsal methods are usually benchmarked on small and simple datasets as they are not powerful enough to generate more complex data with a greater number of classes. We notice that in VAE-based generative replay, this could be attributed to the fact that the generated features are far from the original ones when mapped to the latent space. Therefore, we propose three modifications that allow the model to learn and generate complex data. More specifically, we incorporate the distillation in latent space between the current and previous models to reduce feature drift. Additionally, a latent matching for the reconstruction and original data is proposed to improve generated features alignment. Further, based on the observation that the reconstructions are better for preserving knowledge, we add the cycling of generations through the previously trained model to make them closer to the original data. Our method outperforms other generative replay methods in various scenarios. Code available at https://github.com/valeriya-khan/looking-through-the-past.
Code (1)
Tasks
Continual LearningSimilar Papers 제목 키워드 기반
Continual Learning using the SHDL Framework with Skewed Replay Distributions
Human and animals continuously acquire, adapt as well as transfer knowledge throughout their lifespan. The ability to learn continuously is crucial for the effective functioning of agents interacting with the real world …
Continual LearningM+: Extending MemoryLLM with Scalable Long-Term Memory
Equipping large language models (LLMs) with latent-space memory has attracted increasing attention as they can extend the context window of existing language models. However, retaining information from the distant past r…
16kGPULong-Context UnderstandingText GenerationLLM Unlearning Under the Microscope: A Full-Stack View on Methods and Metrics
Machine unlearning for large language models (LLMs) aims to remove undesired data, knowledge, and behaviors (e.g., for safety, privacy, or copyright) while preserving useful model capabilities. Despite rapid progress ove…
PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective
Parameter-efficient finetuning (PEFT) has become the standard approach for adapting large language models, yet evaluations largely emphasize downstream accuracy while overlooking the retention of pretrained capabilities.…
Oracle-guided Dynamic User Preference Modeling for Sequential Recommendation
Sequential recommendation methods can capture dynamic user preferences from user historical interactions to achieve better performance. However, most existing methods only use past information extracted from user histori…
Sequential Recommendation