paper-with-me

홈 › Papers

Continual Learning in Deep Networks: an Analysis of the Last Layer

2021-06-03 · Timothée Lesort, Thomas George, Irina Rish

We study how different output layer parameterizations of a deep neural network affects learning and forgetting in continual learning settings. The following three effects can cause catastrophic forgetting in the output layer: (1) weights modifications, (2) interference, and (3) projection drift. In this paper, our goal is to provide more insights into how changing the output layer parameterization may address (1) and (2). Some potential solutions to those issues are proposed and evaluated here in several continual learning scenarios. We show that the best-performing type of output layer depends on the data distribution drifts and/or the amount of data available. In particular, in some cases where a standard linear layer would fail, changing parameterization is sufficient to achieve a significantly better performance, without introducing any continual-learning algorithm but instead by using standard SGD to train a model. Our analysis and results shed light on the dynamics of the output layer in continual learning scenarios and suggest a way of selecting the best type of output layer for a given scenario.

📄 PDF Abstract BibTeX arXiv:2106.01834

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

Branch-Tuning: Balancing Stability and Plasticity for Continual Self-Supervised Learning

2024-03-27 · Wenzhuo LIU, Fei Zhu, Cheng-Lin Liu

Self-supervised learning (SSL) has emerged as an effective paradigm for deriving general representations from vast amounts of unlabeled data. However, as real-world applications continually integrate new content, the hig…

Continual LearningContinual Self-Supervised LearningSelf-Supervised Learning

New Insights for the Stability-Plasticity Dilemma in Online Continual Learning

2023-02-17 · Dahuin Jung, Dongjin Lee, Sunwon Hong, Hyemi Jang 외

The aim of continual learning is to learn new tasks continuously (i.e., plasticity) without forgetting previously learned knowledge from old tasks (i.e., stability). In the scenario of online continual learning, wherein …

Continual Learning

Sparsity and Heterogeneous Dropout for Continual Learning in the Null Space of Neural Activations

2022-03-12 · Ali Abbasi, Parsa Nooralinejad, Vladimir Braverman, Hamed Pirsiavash 외

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

NORACL: Neurogenesis for Oracle-free Resource-Adaptive Continual Learning

2026-04-29 · Karthik Charan Raghunathan, Christian Metzner, Laura Kriener, Melika Payvand arxiv

In a continual learning setting, we require a model to be plastic enough to learn a new task and stable enough to not disturb previously learned capabilities. We argue that this dilemma has an architectural root. A finit…

Continual Learning

Statistical mechanics of continual learning: variational principle and mean-field potential

2022-12-06 · Chan Li, Zhenye Huang, Wenxuan Zou, Haiping Huang

An obstacle to artificial general intelligence is set by continual learning of multiple tasks of different nature. Recently, various heuristic tricks, both from machine learning and from neuroscience angles, were propose…

Continual LearningMulti-Task Learning