Continual Learning of Multi-modal Dynamics with External Memory

2022-03-02 · Abdullah Akgül, Gozde Unal, Melih Kandemir

We study the problem of fitting a model to a dynamical environment when new modes of behavior emerge sequentially. The learning model is aware when a new mode appears, but it cannot access the true modes of individual training sequences. The state-of-the-art continual learning approaches cannot handle this setup, because parameter transfer suffers from catastrophic interference and episodic memory design requires the knowledge of the ground-truth modes of sequences. We devise a novel continual learning method that overcomes both limitations by maintaining a \textit{descriptor} of the mode of an encountered sequence in a neural episodic memory. We employ a Dirichlet Process prior on the attention weights of the memory to foster efficient storage of the mode descriptors. Our method performs continual learning by transferring knowledge across tasks by retrieving the descriptors of similar modes of past tasks to the mode of a current sequence and feeding this descriptor into its transition kernel as control input. We observe the continual learning performance of our method to compare favorably to the mainstream parameter transfer approach.

📄 PDF Abstract BibTeX arXiv:2203.00936

Code (1)

adinlab/CDDP-Continual-Learning-of-Multi-modal-Dynamics-with-External-Memory 공식 구현

Tasks

Continual Learning

Methods 이 논문이 사용한 방법론

AWARE We propose to theoretically and empirically examine the effect of incorporating weighting schemes into walk-aggregating GNNs. To this end, we propose a simple, interpretable, and…

Continual Knowledge Updating in LLM Systems: Learning Through Multi-Timescale Memory Dynamics

2026-05-06 · Andreas Pattichis, Constantine Dovrolis arxiv

LLMs are trained once, then deployed into a world that never stops changing. External memory compensates for this, but most systems manage it explicitly rather than letting it adapt on its own. Biological memory works di…

InduceKV: Fixed-Footprint Continual Adaptation of Multimodal LLMs via Inducing KV Memories

2026-07-02 · Qianyu Chen, Ziteng Feng, Canran Xiao, Runxuan Tang arxiv

Multimodal large language models must adapt to evolving tasks and domains, yet continual improvement under bounded deployment footprint remains difficult because repeated parameter updates or growing replay stores can ac…

MemEIC: A Step Toward Continual and Compositional Knowledge Editing

2025-10-29 · Jin Seong, Jiyun Park, Wencke Liermann, Hongseok Choi 외 arxiv

The dynamic nature of information necessitates continuously updating large vision-language models (LVLMs). While recent knowledge editing techniques hint at promising directions, they often focus on editing a single moda…

knowledge editing

When Continual Learning Moves to Memory: A Study of Experience Reuse in LLM Agents

2026-04-29 · Qisheng Hu, Quanyu Long, Wenya Wang arxiv

Memory-augmented LLM agents offer an appealing shortcut to continual learning: rather than updating model parameters, they accumulate experience in external memory, seemingly sidestepping the stability-plasticity dilemma…