Continual Learning with Adaptive Weights (CLAW)

By Tameem Adel, Han Zhao, Richard E. Turner. 2019

79 citations recorded by the source corpus.

Approaches to continual learning aim to successfully learn a set of related tasks that arrive in an online manner. Recently, several frameworks have been developed which enable deep learning to be deployed in this learning scenario. A key modelling decision is to what extent the architecture should be shared across tasks. On the one hand, separately modelling each task avoids catastrophic forgetting but it does not support transfer learning and leads to large models. On the other hand, rigidly specifying a shared component and a task-specific part enables task transfer and limits the model size, but it is vulnerable to catastrophic forgetting and restricts the form of task-transfer that can occur. Ideally, the network should adaptively identify which parts of the network to share in a data driven way. Here we introduce such an approach called Continual Learning with Adaptive Weights (CLAW), which is based on probabilistic modelling and variational inference. Experiments show that CLAW achieves state-of-the-art performance on six benchmarks in terms of overall continual learning performance, as measured by classification accuracy, and in terms of addressing catastrophic forgetting.

Subfields: Artificial Intelligence, Meta & Continual Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Meta & Continual Learning

Citations in the corpus

A Comprehensive Survey of Continual Learning: Theory, Method and Application (2023, background): “The second is online variational inference (VI) [55], [56], [57], [58], [59], [60].”
Adaptive Orthogonal Projection for Batch and Online Continual Learning (2022, background): “Many other papers use related approaches (Zenke, Poole, and Ganguli 2017; Fernando et al. 2017; Aljundi et al. 2017; Ritter, Botev, and Barber 2018; Xu and Zhu 2018; Kemker and Kanan 2018; Parisi et al. 2018; Ahn et al. 2019; Hu et al. 2021; Dhar et al. 2019; Adel, Zhao, and Turner 2020).”
GCR: Gradient Coreset based Replay Buffer Selection for Continual Learning (2021, neutral): “Other work attempts to identify shared information across tasks (CLAW [19]) using variational inference.”
Continual World: A Robotic Benchmark For Continual Reinforcement Learning (2021, methodology): “[2] Tameem Adel, Han Zhao, and Richard E.”
Continual Learning by Using Information of Each Class Holistically (2021, background): “Many papers dealing with CF use a similar idea (Zenke, Poole, and Ganguli 2017; Fernando et al. 2017; Aljundi et al. 2017; Ritter, Botev, and Barber 2018; Xu and Zhu 2018; Kemker and Kanan 2018; Parisi et al. 2018; Dhar et al. 2019; Adel, Zhao, and Turner 2020).”
Training Networks in Null Space of Feature Covariance for Continual Learning (2021, positive): “Under the Bayesian framework, VCL [35], CLAW [2] and IMM [24] take the posterior distribution of network parameters learned from previous tasks as the prior distribution of network parameters on current task, which implicitly penalizes the changes of network parameters.”
Generalized Variational Continual Learning (2020, methodology): “Adel et al. (2020) propose CLAW, which also uses variational inference alongside pertask parameters, but requires a more complex meta-learning based training procedure involving multiple splits of the dataset.”
Embracing Change: Continual Learning in Deep Neural Networks. (2020, background): “Other solutions judiciously add new capacity only when absolutely needed [36 – 38], expand more aggressively and then prune or compress parts of the model [39,40], or channel learning of new tasks to unused parts of a large ﬁ xed model [41 – 45].”
MgSvF: Multi-Grained Slow versus Fast Framework for Few-Shot Class-Incremental Learning (2020, neutral): “Recently, there has been a large body of research in incremental learning [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43].”
Gradient-based Editing of Memory Examples for Online Task-free Continual Learning (2020, neutral): “To mitigate this effect, a wide variety of approaches have been investigated such as adding regularization [1, 19, 30, 43], separating parameters for previous and new data [24, 36, 37], replaying examples from memory or a generative model [26, 33, 38], meta-learning [18].”
Bayesian Structure Adaptation for Continual Learning (2019, background): “One recent approach (Adel et al., 2019), instead of using binary masks, tries to adapts weights at different scales for different tasks; it is also designed only for discriminative tasks.”
Class-Incremental Learning: Survey and Performance Evaluation on Image Classification (2020, neutral): “CLAW [83] extends VCL by applying an attention mechanism on the whole network which allows automation of the architecture adaptation process that assigns parameters to be ﬁxed or not after each task.”
The Effectiveness of Memory Replay in Large Scale Continual Learning (2020, neutral): “(Kirkpatrick et al., 2017; Zenke et al., 2017; Ritter et al., 2018; Fara-jtabar et al., 2020; Yin et al., 2020; Adel et al., 2020).”
How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition (2022, background): “Existing works have achieved this either via variational inference (Adel et al., 2020) or via expectation maximization (Lee et al., 2021a).”
TAG: Task-based Accumulated Gradients for Lifelong learning (2021, methodology): “are approaches based on approximate Bayesian inference, where parameters are sampled from a distribution, that suggest controlling the updates based on parameter uncertainty [6, 1, 2].”
BNS: Building Network Structures Dynamically for Continual Learning (2021, background): “Many papers dealing with CF employ similar ideas [63, 13, 3, 44, 60, 23, 40, 11, 1, 48].”
Gradient Based Memory Editing for Task-Free Continual Learning (2020, background): “Regularization based approaches (Kirk-patrick et al., 2017; Zenke et al., 2017; Nguyen et al., 2018; Adel et al., 2020) discourage the change of parameters that are important to previous data.”
Few-Shot Class-Incremental Learning via Feature Space Composition (2020, background): “Recently, there has been a large body of research in continual learning [1, 2, 19–26].”