A Practical Online Framework for Extracting Running Video Summaries under a Fixed Memory Budget

Lavania, Chandrashekhar; Wei, Kai; Iyer, Rishabh; Bilmes, Jeff

doi:10.1137/1.9781611976700.26

Cited by 1 publication

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the labels of the queried data are not available before the query, which is different to our setting where both the input and target are available. Online submodular maximization (Buchbinder et al, 2014;Lavania et al, 2021) selects a memory buffer online to maximize a submodular criterion. They propose greedy algorithms based on an improvement thresholding procedure, which is similar to our InfoGS.…”

Section: Related Workmentioning

confidence: 99%

Information-theoretic Online Memory Selection for Continual Learning

Sun¹,

Calandriello²,

Hu³

et al. 2022

Preprint

View full text Add to dashboard Cite

A challenging problem in task-free continual learning is the online selection of a representative replay memory from data streams. In this work, we investigate the online memory selection problem from an information-theoretic perspective. To gather the most information, we propose the surprise and the learnability criteria to pick informative points and to avoid outliers. We present a Bayesian model to compute the criteria efficiently by exploiting rank-one matrix structures. We demonstrate that these criteria encourage selecting informative points in a greedy algorithm for online memory selection. Furthermore, by identifying the importance of the timing to update the memory, we introduce a stochastic informationtheoretic reservoir sampler (InfoRS), which conducts sampling among selective points with high information. Compared to reservoir sampling, InfoRS demonstrates improved robustness against data imbalance. Finally, empirical performances over continual learning benchmarks manifest its efficiency and efficacy.

show abstract

Section: Related Workmentioning

confidence: 99%