Don't forget, there is more than forgetting: new metrics for Continual Learning

Díaz-Rodríguez, Natalia; Lomonaco, Vincenzo; Filliat, David; Maltoni, Davide

doi:10.48550/arxiv.1810.13166

Cited by 23 publications

(31 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also consider training time as a significant evaluation factor [11], especially in real-time applications. Besides, algorithm ranking metrics are proposed according to different desiderata in [4], including accuracy, forward/backward transfer, model size efficiency, sample storage size efficiency, computational efficiency. Díaz-Rodríguez et al fuse these desiderata to a single CL score for ranking purposes.…”

Section: Research Questionsmentioning

confidence: 99%

Adaptive Explainable Continual Learning Framework for Regression Problems with Focus on Power Forecasts

2021

Preprint

View full text Add to dashboard Cite

Compared with traditional deep learning techniques, continual learning enables deep neural networks to learn continually and adaptively. Deep neural networks have to learn new tasks and overcome forgetting the knowledge obtained from the old tasks as the amount of data keeps increasing in applications. In this article, two continual learning scenarios will be proposed to describe the potential challenges in this context. Besides, based on our previous work regarding the CLeaR framework, which is short for continual learning for regression tasks, the work will be further developed to enable models to extend themselves and learn data successively. Research topics are related but not limited to developing continual deep learning algorithms, strategies for non-stationarity detection in data streams, explainable and visualizable artificial intelligence, etc. Moreover, the framework-and algorithm-related hyperparameters should be dynamically updated in applications. Forecasting experiments will be conducted based on power generation and consumption data collected from real-world applications. A series of comprehensive evaluation metrics and visualization tools can help analyze the experimental results. The proposed framework is expected to be generally applied to other constantly changing scenarios.

show abstract

Section: Research Questionsmentioning

confidence: 99%

Adaptive Explainable Continual Learning Framework for Regression Problems with Focus on Power Forecasts

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Several studies suggested that forward knowledge transfer is critical for continual learning [17,4], which might be either positive or negative due to the dynamic data distributions. Although it is highly nontrivial to mitigate potential negative transfer while overcoming catastrophic forgetting, the efforts that specifically consider this challenging issue are limited.…”

Section: Related Workmentioning

confidence: 99%

“…(1) CIFAR-100-SC [34]: CIFAR-100 can be split as 20 superclasses (SC) with 5 classes per superclass dependent on semantic similarity, where each superclass is a classification task. Since the superclasses are semantically different, forward knowledge transfer in such a task sequence is Architecture: We follow [10] to use a CNN architecture with 6 convolution layers and 2 fully connected layers for benchmark (1, 2, 3), and AlexNet [14] for benchmark (4,5). Since continual learning needs to quickly learn a usable model from incrementally collected data, we mainly consider learning the network from scratch.…”

Section: Visual Classification Tasksmentioning

confidence: 99%

“…3 For P&C [26], we make a grid search of the hyperparameter that controls the EWC penalty. 4 Since β is integrated into λ e , AFEC only needs to make a grid search of λ e while keeping λ the same as the corresponding weight regularization approaches. We follow the implementation of [10] for visual classification tasks with small-scale and large-scale images.…”

Section: B3 New Log Posterior Of Afecmentioning

confidence: 99%

See 1 more Smart Citation

AFEC: Active Forgetting of Negative Transfer in Continual Learning

Wang¹,

Zhang²,

Jia³

et al. 2021

Preprint

View full text Add to dashboard Cite

Continual learning aims to learn a sequence of tasks from dynamic data distributions. Without accessing to the old training samples, knowledge transfer from the old tasks to each new task is difficult to determine, which might be either positive or negative. If the old knowledge interferes with the learning of a new task, i.e., the forward knowledge transfer is negative, then precisely remembering the old tasks will further aggravate the interference, thus decreasing the performance of continual learning. By contrast, biological neural networks can actively forget the old knowledge that conflicts with the learning of a new experience, through regulating the learning-triggered synaptic expansion and synaptic convergence. Inspired by the biological active forgetting, we propose to actively forget the old knowledge that limits the learning of new tasks to benefit continual learning. Under the framework of Bayesian continual learning, we develop a novel approach named Active Forgetting with synaptic Expansion-Convergence (AFEC). Our method dynamically expands parameters to learn each new task and then selectively combines them, which is formally consistent with the underlying mechanism of biological active forgetting. We extensively evaluate AFEC on a variety of continual learning benchmarks, including CIFAR-10 regression tasks, visual classification tasks and Atari reinforcement tasks, where AFEC effectively improves the learning of new tasks and achieves the state-of-the-art performance in a plug-and-play way.

show abstract

“…However, the broad range of continual learning methods remain largely unexplored for knowledge graph embedding. Furthermore, the implications of assumptions in the context of robotics is not well documented due to the definition of different task specific measures and a focus on the final inference performance [15].…”

Section: Introductionmentioning

confidence: 99%

Continual Learning of Knowledge Graph Embeddings

Daruna¹,

Gupta²,

Sridharan³

et al. 2021

Preprint

View full text Add to dashboard Cite

In recent years, there has been a resurgence in methods that use distributed (neural) representations to represent and reason about semantic knowledge for robotics applications. However, while robots often observe previously unknown concepts, these representations typically assume that all concepts are known a priori, and incorporating new information requires all concepts to be learned afresh. Our work relaxes the static assumptions of these representations to tackle the incremental knowledge graph embedding problem by leveraging principles of a range of continual learning methods. Through an experimental evaluation with several knowledge graphs and embedding representations, we provide insights about trade-offs for practitioners to match a semantics-driven robotics applications to a suitable continual knowledge graph embedding method.

show abstract

Don't forget, there is more than forgetting: new metrics for Continual Learning

Cited by 23 publications

References 9 publications

Adaptive Explainable Continual Learning Framework for Regression Problems with Focus on Power Forecasts

Adaptive Explainable Continual Learning Framework for Regression Problems with Focus on Power Forecasts

AFEC: Active Forgetting of Negative Transfer in Continual Learning

Continual Learning of Knowledge Graph Embeddings

Contact Info

Product

Resources

About