Storage Fit Learning with Feature Evolvable Streams

07/22/2020
by   Bo-Jian Hou, et al.
0

Feature evolvable learning has been widely studied in recent years where old features will vanish and new features will emerge when learning with streams. Conventional methods usually assume that a label will be revealed after prediction at each time step. However, in practice, this assumption may not hold whereas no label will be given at most time steps. To tackle this problem, we leverage the technique of manifold regularization to utilize the previous similar data to assist the refinement of the online model. Nevertheless, this approach needs to store all previous data which is impossible in learning with streams that arrive sequentially in large volume. Thus we need a buffer to store part of them. Considering that different devices may have different storage budgets, the learning approaches should be flexible subject to the storage budget limit. In this paper, we propose a new setting: Storage-Fit Feature-Evolvable streaming Learning (SF2EL) which incorporates the issue of rarely-provided labels into feature evolution. Our framework is able to fit its behavior to different storage budgets when learning with feature evolvable streams with unlabeled data. Besides, both theoretical and empirical results validate that our approach can preserve the merit of the original feature evolvable learning i.e., can always track the best baseline and thus perform well at any time step.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/16/2017

Learning with Feature Evolvable Streams

Learning with streaming data has attracted much attention during the pas...
research
04/27/2019

Prediction with Unpredictable Feature Evolution

Feature space can change or evolve when learning with streaming data. Se...
research
11/17/2019

Rebalancing Learning on Evolving Data Streams

Nowadays, every device connected to the Internet generates an ever-growi...
research
04/25/2022

Online Deep Learning from Doubly-Streaming Data

This paper investigates a new online learning problem with doubly-stream...
research
07/03/2020

CICLAD: A Fast and Memory-efficient Closed Itemset Miner for Streams

Mining association rules from data streams is a challenging task due to ...
research
06/07/2021

Enabling On-Device Self-Supervised Contrastive Learning With Selective Data Contrast

After a model is deployed on edge devices, it is desirable for these dev...
research
11/30/2022

Continual Learning with Optimal Transport based Mixture Model

Online Class Incremental learning (CIL) is a challenging setting in Cont...

Please sign up or login with your details

Forgot password? Click here to reset