Coresets via Bilevel Optimization for Continual Learning and Streaming

by   Zalán Borsos, et al.

Coresets are small data summaries that are sufficient for model training. They can be maintained online, enabling efficient handling of large data streams under resource constraints. However, existing constructions are limited to simple models such as k-means and logistic regression. In this work, we propose a novel coreset construction via cardinality-constrained bilevel optimization. We show how our framework can efficiently generate coresets for deep neural networks, and demonstrate its empirical benefits in continual learning and in streaming settings.



page 2

page 7

page 20


Data Summarization via Bilevel Optimization

The increasing availability of massive data sets poses a series of chall...

Online Continual Learning on Class Incremental Blurry Task Configuration with Anytime Inference

Despite rapid advances in continual learning, a large body of research i...

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

Compared with traditional deep learning techniques, continual learning e...

One Pass ImageNet

We present the One Pass ImageNet (OPIN) problem, which aims to study the...

Continual Learning by Asymmetric Loss Approximation with Single-Side Overestimation

Catastrophic forgetting is a critical challenge in training deep neural ...

Continual learning improves Internet video streaming

We describe Fugu, a continual learning algorithm for bitrate selection i...

Maximum likelihood estimation of a finite mixture of logistic regression models in a continuous data stream

In marketing we are often confronted with a continuous stream of respons...

Code Repositories


Coresets via Bilevel Optimization

view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.