Incremental Feature Learning For Infinite Data

TL;DR

This paper proposes an adaptive incremental learning method for credit-card fraud detection that dynamically adjusts neural network architecture for each data chunk, improving accuracy without retaining all data.

Contribution

It introduces a novel incremental feature learning paradigm that dynamically determines optimal network architecture for each data chunk, enhancing fraud detection accuracy.

Findings

Demonstrates superior accuracy on real fraud dataset

Effectively adapts to new data chunks without storing all data

Outperforms fixed-architecture incremental methods

Abstract

This study addresses the actual behavior of the credit-card fraud detection environment where financial transactions containing sensitive data must not be amassed in an enormous amount to conduct learning. We introduce a new adaptive learning approach that adjusts frequently and efficiently to new transaction chunks; each chunk is discarded after each incremental training step. Our approach combines transfer learning and incremental feature learning. The former improves the feature relevancy for subsequent chunks, and the latter, a new paradigm, increases accuracy during training by determining the optimal network architecture dynamically for each new chunk. The architectures of past incremental approaches are fixed; thus, the accuracy may not improve with new chunks. We show the effectiveness and superiority of our approach experimentally on an actual fraud dataset.