Lifelong machine learning: a paradigm for continuous learning

By Zhiyuan Chen, B. Liu. 2017

333 citations recorded by the source corpus.

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

1. Bing Liu
2. Zixuan Ke
3. Hu Xu
4. Lei Shu
5. Bernt Schiele
6. Sergey Levine
7. Trevor Darrell
8. Han-Jia Ye

In Meta & Continual Learning

1. Bing Liu
2. Zixuan Ke
3. Hu Xu
4. Bin Liu
5. Gyuhak Kim
6. Lei Shu
7. Jorge Armando Mendez Mendez
8. Eric Eaton

Citations in the corpus

• Learnability and Algorithm for Continual Learning (2023, background): “Learning a sequence of tasks incrementally, called continual learning, has attracted a great deal of attention recently (Chen & Liu, 2018).”
• CoMPS: Continual Meta Policy Search (2021, background): “Continual or online learning studies the streaming data setting, where experience is used for training as soon as it is received (Thrun & Pratt, 2012; Hazan et al., 2016; Chen & Liu, 2016; Parisi et al., 2019; Khetarpal et al., 2020).”
• Detection and Continual Learning of Novel Face Presentation Attacks (2021, methodology): “Recent works of CL [2] for deep neural networks have mostly focused on tackling catastrophic forgetting [3].”
• Online Continual Learning in Image Classification: An Empirical Survey (2021, methodology): “Since CL is often used interchangeably with lifelong learning [6, 7] and incremental learning [8, 9], for simplicity, we will use CL to refer to all concepts mentioned above.”
• Curriculum-Meta Learning for Order-Robust Continual Relation Extraction (2021, background): “Catastrophic Forgetting (CF) is a well-known problem in continual learning (Chen and Liu 2016).”
• Adaptive Aggregation Networks for Class-Incremental Learning (2020, background): “Closely related topics are referred to as continual learning [10, 26] and lifelong learning [2, 7, 22].”
• Continual Learning with Knowledge Transfer for Sentiment Classification (2021, background): “Continual learning (CL) aims to learn a sequence of tasks incrementally [4, 18].”
• Rehearsal-Free Continual Learning over Small Non-I.I.D. Batches (2019, positive): “on regularization, architectural or rehearsal approaches [24, 17, 2].”
• Few-Shot Object Detection via Feature Reweighting (2018, background): “This also opens up our model’s potential in a life-long learning setting [5], where the model accumulates the knowledge learned from past and uses/adapts it for future prediction.”
• Lifelong Learning CRF for Supervised Aspect Extraction (2017, methodology): “Due to leveraging the knowledge gained from the past to help the new domain extraction, we are using the idea of lifelong machine learning (LML) (Chen and Liu, 2016; Thrun, 1998; Silver et al., 2013), which is a continuous learning paradigm that retains the knowledge learned in the past and uses it ”
• CLASSIC: Continual and Contrastive Learning of Aspect Sentiment Classification Tasks (2021, background): “After learning a task, its training data is often discarded (Chen and Liu, 2018).”
• Learning Transferable Knowledge Through Embedding Spaces (2019, neutral): “Similar to the previous chapter, we consider a lifelong learning setting [214], where a learning agent faces multiple, consecutive tasks {Z(t)}Max t=1 in a sequence t = 1, .”
• Overcoming Catastrophic Forgetting for Continual Learning via Model Adaptation (2018, background): “Many approaches have been proposed to deal with catastrophic forgetting (CF), which is one of the challenging problems of neural networks for lifelong learning (Chen & Liu, 2018).”
• Deep Bayesian Unsupervised Lifelong Learning (2021, background): “Currently, there are no universal past knowledge preservation schemes for different algorithms and settings [13] in LL.”
• TAG: Task-based Accumulated Gradients for Lifelong learning (2021, background): “Unlike human beings, who can efficiently assess the correctness and applicability of the past knowledge [12], neural networks and other machine learning models often face various issues in this setting.”
• Sentiment Analysis Based on Deep Learning: A Comparative Study (2020, methodology): “tering [57]. In addition, one advanced machine learning paradigm is the so-called holistic models or lifelong learning algorithms, which are argued to significantly improve sentiment analysis accuracy [58]. While other methods learn a model by using only data for a particular application, this method”