Online Class-Incremental Continual Learning with Adversarial Shapley Value

Shim, Dongsub; Mai, Zheda; Jeong, Jihwan; Sanner, Scott; Kim, Hyun Woo; Jang, Ji Hye

doi:10.1609/aaai.v35i11.17159

Cited by 113 publications

(41 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alternatively, some sampling methods focus on better memory retrieval strategies that reduce forgetting, e.g. MIR [2], ASER [42], and GMED [22]. (iii) In other approaches, GDumb [36] proposed a degenerate solution to the problem of online learning ignoring the stream data and learning only on the memory samples.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Evaluation in Online Continual Learning: A New Hope

Yasir¹,

Bibi²,

Alhamoud³

et al. 2023

Preprint

View full text Add to dashboard Cite

Current evaluations of Continual Learning (CL) methods typically assume that there is no constraint on training time and computation. This is an unrealistic assumption for any real-world setting, which motivates us to propose: a practical real-time evaluation of continual learning, in which the stream does not wait for the model to complete training before revealing the next data for predictions. To do this, we evaluate current CL methods with respect to their computational costs. We hypothesize that under this new evaluation paradigm, computationally demanding CL approaches may perform poorly on streams with a varying distribution. We conduct extensive experiments on CLOC, a large-scale dataset containing 39 million time-stamped images with geolocation labels. We show that a simple baseline outperforms state-of-the-art CL methods under this evaluation, questioning the applicability of existing methods in realistic settings. In addition, we explore various CL components commonly used in the literature, including memory sampling strategies and regularization approaches. We find that all considered methods fail to be competitive against our simple baseline. This surprisingly suggests that the majority of existing CL literature is tailored to a specific class of streams that is not practical. We hope that the evaluation we provide will be the first step towards a paradigm shift to consider the computational cost in the development of online continual learning methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Evaluation in Online Continual Learning: A New Hope

Yasir¹,

Bibi²,

Alhamoud³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, Herding [44], K-Means [15], OCS [57], InfoRS [49], RM [6], and GSS [5] aim to maximize diversity among samples selected for training with different metrics. MIR [3], ASER [48], and CLIB [27] rank the samples according to their informativeness and select the top-k. Lastly, balanced sampling [16,17,40] select samples such that an equal distribution of classes is selected for training. In our experiments, we only consider previous sampling strategies that are applicable to our setup and compare them against Naive.…”

Section: Dissecting Continual Learning Systemsmentioning

confidence: 99%

Computationally Budgeted Continual Learning: What Does Matter?

Prabhu¹,

Hammoud²,

Dokania³

et al. 2023

Preprint

View full text Add to dashboard Cite

Continual Learning (CL) aims to sequentially train models on streams of incoming data that vary in distribution by preserving previous knowledge while adapting to new data. Current CL literature focuses on restricted access to previously seen data, while imposing no constraints on the computational budget for training. This is unreasonable for applications in-the-wild, where systems are primarily constrained by computational and time budgets, not storage. We revisit this problem with a large-scale benchmark and analyze the performance of traditional CL approaches in a compute-constrained setting, where effective memory samples used in training can be implicitly restricted as a consequence of limited computation. We conduct experiments evaluating various CL sampling strategies, distillation losses, and partial fine-tuning on two large-scale datasets, namely ImageNet2K and Continual Google Landmarks V2 in data incremental, class incremental, and time incremental settings. Through extensive experiments amounting to a total of over 1500 GPU-hours, we find that, under computeconstrained setting, traditional CL approaches, with no exception, fail to outperform a simple minimal baseline that samples uniformly from memory. Our conclusions are consistent in a different number of stream time steps, e.g., 20 to 200, and under several computational budgets. This suggests that most existing CL methods are particularly too computationally expensive for realistic budgeted deployment. Code for this project is available at: https: //github.com/drimpossible/BudgetCL.

show abstract

“…Similarity learning has emerged as a significant area of machine learning research with many real-world applications [11][12][13][14]. Most current research focuses on preserving exemplars of previous classes to remind the network of past knowledge by finding exemplars that are useful for knowledge retention [15][16][17]. Recent advances include techniques such as using supervised contrastive learning with nearest-class-mean to maximize the use of exemplars [18].…”

Section: Introductionmentioning

confidence: 99%

Incremental class learning using variational autoencoders with similarity learning

Huo

Zyl

2023

Neural Comput & Applic

View full text Add to dashboard Cite

Catastrophic forgetting in neural networks during incremental learning remains a challenging problem. Previous research investigated catastrophic forgetting in fully connected networks, with some earlier work exploring activation functions and learning algorithms. Applications of neural networks have been extended to include similarity learning. Understanding how similarity learning loss functions would be affected by catastrophic forgetting is of significant interest. Our research investigates catastrophic forgetting for four well-known similarity-based loss functions during incremental class learning. The loss functions are Angular, Contrastive, Center, and Triplet loss. Our results show that the catastrophic forgetting rate differs across loss functions on multiple datasets. The Angular loss was least affected, followed by Contrastive, Triplet loss, and Center loss with good mining techniques. We implemented three existing incremental learning techniques, iCaRL, EWC, and EBLL. We further proposed a novel technique using Variational Autoencoders (VAEs) to generate representation as exemplars passed through the network’s intermediate layers. Our method outperformed three existing state-of-the-art techniques. We show that one does not require stored images (exemplars) for incremental learning with similarity learning. The generated representations from VAEs help preserve regions of the embedding space used by prior knowledge so that new knowledge does not “overwrite” it.

show abstract

Online Class-Incremental Continual Learning with Adversarial Shapley Value

Cited by 113 publications

References 13 publications

Real-Time Evaluation in Online Continual Learning: A New Hope

Real-Time Evaluation in Online Continual Learning: A New Hope

Computationally Budgeted Continual Learning: What Does Matter?

Incremental class learning using variational autoencoders with similarity learning

Contact Info

Product

Resources

About