How Efficient Are Today's Continual Learning Algorithms?

• URL: http://arxiv.org/abs/2303.18171v2
• Date: Mon, 3 Apr 2023 13:53:33 GMT
• Title: How Efficient Are Today's Continual Learning Algorithms?
• Authors: Md Yousuf Harun, Jhair Gallardo, Tyler L. Hayes, Christopher Kanan
• Abstract summary: Supervised continual learning involves updating a deep neural network (DNN) from an ever-growing stream of labeled data. One of the major motivations behind continual learning is being able to efficiently update a network with new information, rather than retraining from scratch on the training dataset as it grows over time. Here, we study recent methods for incremental class learning and illustrate that many are highly inefficient in terms of compute, memory, and storage.
• Score: 31.120016345185217
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Supervised Continual learning involves updating a deep neural network (DNN) from an ever-growing stream of labeled data. While most work has focused on overcoming catastrophic forgetting, one of the major motivations behind continual learning is being able to efficiently update a network with new information, rather than retraining from scratch on the training dataset as it grows over time. Despite recent continual learning methods largely solving the catastrophic forgetting problem, there has been little attention paid to the efficiency of these algorithms. Here, we study recent methods for incremental class learning and illustrate that many are highly inefficient in terms of compute, memory, and storage. Some methods even require more compute than training from scratch! We argue that for continual learning to have real-world applicability, the research community cannot ignore the resources used by these algorithms. There is more to continual learning than mitigating catastrophic forgetting.

Related papers

• Continual Learning of Numerous Tasks from Long-tail Distributions [17.706669222987273]
  Continual learning focuses on developing models that learn and adapt to new tasks while retaining previously acquired knowledge. Existing continual learning algorithms usually involve a small number of tasks with uniform sizes and may not accurately represent real-world learning scenarios. We propose a method that reuses the states in Adam by maintaining a weighted average of the second moments from previous tasks. We demonstrate that our method, compatible with most existing continual learning algorithms, effectively reduces forgetting with only a small amount of additional computational or memory costs.
  arXiv  Detail & Related papers  (2024-04-03T13:56:33Z)
• Accelerating Batch Active Learning Using Continual Learning Techniques [5.514154947870141]
  A major problem with Active Learning (AL) is high training costs since models are typically retrained from scratch after every query round. We develop a new class of techniques, circumventing this problem, by biasing further training towards previously labeled sets. We conduct experiments across many data domains, including natural language, vision, medical imaging, and computational biology.
  arXiv  Detail & Related papers  (2023-05-10T18:34:05Z)
• PIVOT: Prompting for Video Continual Learning [50.80141083993668]
  We introduce PIVOT, a novel method that leverages extensive knowledge in pre-trained models from the image domain. Our experiments show that PIVOT improves state-of-the-art methods by a significant 27% on the 20-task ActivityNet setup.
  arXiv  Detail & Related papers  (2022-12-09T13:22:27Z)
• Continual Learning with Deep Learning Methods in an Application-Oriented Context [0.0]
  An important research area of Artificial Intelligence (AI) deals with the automatic derivation of knowledge from data. One type of machine learning algorithms that can be categorized as "deep learning" model is referred to as Deep Neural Networks (DNNs) DNNs are affected by a problem that prevents new knowledge from being added to an existing base.
  arXiv  Detail & Related papers  (2022-07-12T10:13:33Z)
• Explain to Not Forget: Defending Against Catastrophic Forgetting with XAI [10.374979214803805]
  Catastrophic forgetting describes the phenomenon when a neural network completely forgets previous knowledge when given new information. We propose a novel training algorithm called training by explaining in which we leverage Layer-wise Relevance propagation in order to retain the information a neural network has already learned in previous tasks when training on new data. Our method not only successfully retains the knowledge of old tasks within the neural networks but does so more resource-efficiently than other state-of-the-art solutions.
  arXiv  Detail & Related papers  (2022-05-04T08:00:49Z)
• On Generalizing Beyond Domains in Cross-Domain Continual Learning [91.56748415975683]
  Deep neural networks often suffer from catastrophic forgetting of previously learned knowledge after learning a new task. Our proposed approach learns new tasks under domain shift with accuracy boosts up to 10% on challenging datasets such as DomainNet and OfficeHome.
  arXiv  Detail & Related papers  (2022-03-08T09:57:48Z)
• Online Continual Learning with Natural Distribution Shifts: An Empirical Study with Visual Data [101.6195176510611]
  "Online" continual learning enables evaluating both information retention and online learning efficacy. In online continual learning, each incoming small batch of data is first used for testing and then added to the training set, making the problem truly online. We introduce a new benchmark for online continual visual learning that exhibits large scale and natural distribution shifts.
  arXiv  Detail & Related papers  (2021-08-20T06:17:20Z)
• Evolving Reinforcement Learning Algorithms [186.62294652057062]
  We propose a method for meta-learning reinforcement learning algorithms. The learned algorithms are domain-agnostic and can generalize to new environments not seen during training. We highlight two learned algorithms which obtain good generalization performance over other classical control tasks, gridworld type tasks, and Atari games.
  arXiv  Detail & Related papers  (2021-01-08T18:55:07Z)
• Mastering Rate based Curriculum Learning [78.45222238426246]
  We argue that the notion of learning progress itself has several shortcomings that lead to a low sample efficiency for the learner. We propose a new algorithm, based on the notion of mastering rate, that significantly outperforms learning progress-based algorithms.
  arXiv  Detail & Related papers  (2020-08-14T16:34:01Z)
• Bilevel Continual Learning [76.50127663309604]
  We present a novel framework of continual learning named "Bilevel Continual Learning" (BCL) Our experiments on continual learning benchmarks demonstrate the efficacy of the proposed BCL compared to many state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-07-30T16:00:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.