iTAML: An Incremental Task-Agnostic Meta-learning Approach

Rajasegaran, Jathushan; Khan, Salman; Hayat, Munawar; Khan, Fahad Shahbaz; Shah, Mubarak

doi:10.1109/cvpr42600.2020.01360

Cited by 121 publications

(105 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with other few-shot detection algorithms, the advantage of ONCE is that after training on the basic dataset, the new small sample dataset can be directly used for inference, and the contents of the basic dataset will not be forgotten in this process. iTAML [ 118 ] is also an incremental learning algorithm designed based on meta-learning, but it focuses on solving classification tasks.…”

Section: Methods Descriptionmentioning

confidence: 99%

An Appraisal of Incremental Learning Methods

Luo

Yin

Bai

et al. 2020

Entropy

View full text Add to dashboard Cite

As a special case of machine learning, incremental learning can acquire useful knowledge from incoming data continuously while it does not need to access the original data. It is expected to have the ability of memorization and it is regarded as one of the ultimate goals of artificial intelligence technology. However, incremental learning remains a long term challenge. Modern deep neural network models achieve outstanding performance on stationary data distributions with batch training. This restriction leads to catastrophic forgetting for incremental learning scenarios since the distribution of incoming data is unknown and has a highly different probability from the old data. Therefore, a model must be both plastic to acquire new knowledge and stable to consolidate existing knowledge. This review aims to draw a systematic review of the state of the art of incremental learning methods. Published reports are selected from Web of Science, IEEEXplore, and DBLP databases up to May 2020. Each paper is reviewed according to the types: architectural strategy, regularization strategy and rehearsal and pseudo-rehearsal strategy. We compare and discuss different methods. Moreover, the development trend and research focus are given. It is concluded that incremental learning is still a hot research area and will be for a long period. More attention should be paid to the exploration of both biological systems and computational models.

show abstract

Section: Methods Descriptionmentioning

confidence: 99%

An Appraisal of Incremental Learning Methods

Luo

Yin

Bai

et al. 2020

Entropy

View full text Add to dashboard Cite

show abstract

“…This simple method, known as experience replay, has been explored and shown to be effective. [5][6][7][8]11,27,50 In this work we aim to go one step further and investigate the role of explanations in continual learning, particularly on mitigating forgetting and change of model explanations.…”

Section: Remembering For the Right Reasonsmentioning

confidence: 99%

“…An active line of research in continual learning explores the effectiveness of using small memory budgets to store data points from the training set, [5][6][7][8] gradients, 9 or storing an online generative model that can fake them later. 10 Memory has been also exploited in the form of accommodating space for architecture growth and storage to fully recover the old performance when needed.…”

Section: Introductionmentioning

confidence: 99%

“…(right) In our proposed idea (RRR), in addition to ℳ rep , we also store model explanations (saliency maps) as ℳ RRR for those samples and encourage the model to remember the original reasoning for the prediction. Note that the saliency maps are small masks resulting in a negligible memory overhead (see section 5.1) state of the art class incremental learning (CIL) methods, including iTAML, 6 EEIl, 5 BiC, 8 TOPIC, 23 iCaRL, 7 EWC, 24 and LwF. 13 Note that RRR does not require task IDs at test time.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Remembering for the right reasons: Explanations reduce catastrophic forgetting

et al. 2021

View full text Add to dashboard Cite

The goal of continual learning (CL) is to learn a sequence of tasks without suffering from the phenomenon of catastrophic forgetting. Previous work has shown that leveraging memory in the form of a replay buffer can reduce performance degradation on prior tasks. We hypothesize that forgetting can be further reduced when the model is encouraged to remember the evidence for previously made decisions. As a first step towards exploring this hypothesis, we propose a simple novel training paradigm, called Remembering for the Right Reasons (RRR), that additionally stores visual model explanations for each example in the buffer and ensures the model has "the right reasons" for its predictions by encouraging its explanations to remain consistent with those used to make decisions at training time. Without this constraint, there is a drift in explanations and increase in forgetting as conventional continual learning algorithms learn new tasks. We demonstrate how RRR can be easily added to any memory or regularization-based approach and results in reduced forgetting, and more importantly, improved model explanations. We have evaluated our approach in the standard and fewshot settings and observed a consistent improvement across various CL approaches using different architectures and techniques to generate model explanations and demonstrated our approach showing a promising connection between explainability and continual learning. Our code is available at https://github.com/ SaynaEbrahimi/Remembering-for-the-Right-Reasons.

show abstract

“…In effect, the ability to generalize across tasks is at the core of metalearning. In this context, [23], [24], [44] propose different metalearning strategies to tackle a continual learning scenario. The resulting techniques reduce task interference by avoiding conflicts between current and future gradient directions to update weights.…”

Section: B Meta-learningmentioning

confidence: 99%

Overcoming Catastrophic Forgetting Using Sparse Coding and Meta Learning

2021

View full text Add to dashboard Cite

Continuous learning occurs naturally in human beings. However, Deep Learning methods suffer from a problem known as Catastrophic Forgetting (CF) that consists of a model drastically decreasing its performance on previously learned tasks when it is sequentially trained on new tasks. This situation, known as task interference, occurs when a network modifies relevant weight values as it learns a new task. In this work, we propose two main strategies to face the problem of task interference in convolutional neural networks. First, we use a sparse coding technique to adaptively allocate model capacity to different tasks avoiding interference between them. Specifically, we use a strategy based on group sparse regularization to specialize groups of parameters to learn each task. Afterward, by adding binary masks, we can freeze these groups of parameters, using the rest of the network to learn new tasks. Second, we use a meta learning technique to foster knowledge transfer among tasks, encouraging weight reusability instead of overwriting. Specifically, we use an optimization strategy based on episodic training to foster learning weights that are expected to be useful to solve future tasks. Together, these two strategies help us to avoid interference by preserving compatibility with previous and future weight values. Using this approach, we achieve stateof-the-art results on popular benchmarks used to test techniques to avoid CF. In particular, we conduct an ablation study to identify the contribution of each component of the proposed method, demonstrating its ability to avoid retroactive interference with previous tasks and to promote knowledge transfer to future tasks.

show abstract

iTAML: An Incremental Task-Agnostic Meta-learning Approach

Cited by 121 publications

References 16 publications

An Appraisal of Incremental Learning Methods

An Appraisal of Incremental Learning Methods

Remembering for the right reasons: Explanations reduce catastrophic forgetting

Overcoming Catastrophic Forgetting Using Sparse Coding and Meta Learning

Contact Info

Product

Resources

About