English
Related papers

Related papers: VERSE: Virtual-Gradient Aware Streaming Lifelong L…

200 papers

Despite rapid advancements in lifelong learning (LLL) research, a large body of research mainly focuses on improving the performance in the existing \textit{static} continual learning (CL) setups. These methods lack the ability to succeed…

Machine Learning · Computer Science 2023-01-30 Soumya Banerjee , Vinay Kumar Verma , Vinay P. Namboodiri

When an agent acquires new information, ideally it would immediately be capable of using that information to understand its environment. This is not possible using conventional deep neural networks, which suffer from catastrophic forgetting…

Machine Learning · Computer Science 2020-04-20 Tyler L. Hayes , Christopher Kanan

Continual Learning, also known as Lifelong Learning, aims to continually learn from new data as it becomes available. While prior research on continual learning in automatic speech recognition has focused on the adaptation of models across…

Machine Learning · Computer Science 2022-07-13 Muqiao Yang , Ian Lane , Shinji Watanabe

In lifelong learning, data are used to improve performance not only on the present task, but also on past and future (unencountered) tasks. While typical transfer learning algorithms can improve performance on future tasks, their…

Continual learning is the problem of learning and retaining knowledge through time over multiple tasks and environments. Research has primarily focused on the incremental classification setting, where new tasks/classes are added at discrete…

Machine Learning · Computer Science 2021-09-23 Zhipeng Cai , Ozan Sener , Vladlen Koltun

Continual learning is increasingly sought after in real world machine learning applications, as it enables learning in a more human-like manner. Conventional machine learning approaches fail to achieve this, as incrementally updating the…

Computer Vision and Pattern Recognition · Computer Science 2023-10-31 Joe Khawand , Peter Hanappe , David Colliaux

Humans and animals have the ability to continually acquire, fine-tune, and transfer knowledge and skills throughout their lifespan. This ability, referred to as lifelong learning, is mediated by a rich set of neurocognitive mechanisms that…

Machine Learning · Computer Science 2019-02-12 German I. Parisi , Ronald Kemker , Jose L. Part , Christopher Kanan , Stefan Wermter

Lifelong learning is a long-standing aim for artificial agents that act in dynamic environments, in which an agent needs to accumulate knowledge incrementally without forgetting previously learned representations. We investigate methods for…

Machine Learning · Computer Science 2021-11-17 Vadym Gryshchuk , Cornelius Weber , Chu Kiong Loo , Stefan Wermter

Continual Learning (CL) and Streaming Machine Learning (SML) study the ability of agents to learn from a stream of non-stationary data. Despite sharing some similarities, they address different and complementary challenges. While SML…

Attaining prototypical features to represent class distributions is well established in representation learning. However, learning prototypes online from streaming data proves a challenging endeavor as they rapidly become outdated, caused…

Computer Vision and Pattern Recognition · Computer Science 2021-10-20 Matthias De Lange , Tinne Tuytelaars

This work demonstrates that training autoregressive video diffusion models from a single video stream$\unicode{x2013}$resembling the experience of embodied agents$\unicode{x2013}$is not only possible, but can also be as effective as…

Computer Vision and Pattern Recognition · Computer Science 2025-07-02 Jason Yoo , Yingchen He , Saeid Naderiparizi , Dylan Green , Gido M. van de Ven , Geoff Pleiss , Frank Wood

Lifelong learning can be viewed as a continuous transfer learning procedure over consecutive tasks, where learning a given task depends on accumulated knowledge --- the so-called knowledge base. Most published work on lifelong learning…

Machine Learning · Statistics 2018-10-30 Changjian Shui , Ihsen Hedhli , Christian Gagné

Understanding continuous video streams plays a fundamental role in real-time applications including embodied AI and autonomous driving. Unlike offline video understanding, streaming video understanding requires the ability to process video…

Computer Vision and Pattern Recognition · Computer Science 2025-07-23 Yibin Yan , Jilan Xu , Shangzhe Di , Yikun Liu , Yudi Shi , Qirui Chen , Zeqian Li , Yifei Huang , Weidi Xie

Lifelong learning in artificial intelligence (AI) aims to mimic the biological brain's ability to continuously learn and retain knowledge, yet it faces challenges such as catastrophic forgetting. Recent neuroscience research suggests that…

Artificial Intelligence · Computer Science 2024-09-24 Jin Du , Xinhe Zhang , Hao Shen , Xun Xian , Ganghua Wang , Jiawei Zhang , Yuhong Yang , Na Li , Jia Liu , Jie Ding

Devising intelligent agents able to live in an environment and learn by observing the surroundings is a longstanding goal of Artificial Intelligence. From a bare Machine Learning perspective, challenges arise when the agent is prevented…

Computer Vision and Pattern Recognition · Computer Science 2022-04-27 Matteo Tiezzi , Simone Marullo , Lapo Faggi , Enrico Meloni , Alessandro Betti , Stefano Melacci

Online continual learning aims to get closer to a live learning experience by learning directly on a stream of data with temporally shifting distribution and by storing a minimum amount of data from that stream. In this empirical…

Self-supervised learning (SSL) aims to eliminate one of the major bottlenecks in representation learning - the need for human annotations. As a result, SSL holds the promise to learn representations from data in-the-wild, i.e., without the…

Computer Vision and Pattern Recognition · Computer Science 2022-03-29 Senthil Purushwalkam , Pedro Morgado , Abhinav Gupta

Artificial neural networks have exceeded human-level performance in accomplishing several individual tasks (e.g. voice recognition, object recognition, and video games). However, such success remains modest compared to human intelligence…

Machine Learning · Computer Science 2019-10-21 Rahaf Aljundi

Lifelong machine learning is a novel machine learning paradigm which can continually accumulate knowledge during learning. The knowledge extracting and reusing abilities enable the lifelong machine learning to solve the related problems.…

Computation and Language · Computer Science 2019-06-03 Xianbin Hong , Gautam Pal , Sheng-Uei Guan , Prudence Wong , Dawei Liu , Ka Lok Man , Xin Huang

The problem of catastrophic forgetting occurs in deep learning models trained on multiple databases in a sequential manner. Recently, generative replay mechanisms (GRM), have been proposed to reproduce previously learned knowledge aiming to…

Computer Vision and Pattern Recognition · Computer Science 2020-07-21 Fei Ye , Adrian G. Bors
‹ Prev 1 2 3 10 Next ›