CCL4Rec: Contrast over Contrastive Learning for Micro-video Recommendation

TL;DR

This paper introduces CCL4Rec, a contrastive learning framework that enhances micro-video recommendation by adaptively contrasting augmented user behaviors, effectively filtering noise and improving performance and efficiency.

Contribution

The paper proposes a novel contrast over contrastive learning framework with hardness-aware augmentations for better user representation in recommendation systems.

Findings

CCL4Rec achieves comparable accuracy to state-of-the-art methods.

It significantly improves training and inference speed.

The framework effectively filters noisy user behaviors.

Abstract

Micro-video recommender systems suffer from the ubiquitous noises in users' behaviors, which might render the learned user representation indiscriminating, and lead to trivial recommendations (e.g., popular items) or even weird ones that are far beyond users' interests. Contrastive learning is an emergent technique for learning discriminating representations with random data augmentations. However, due to neglecting the noises in user behaviors and treating all augmented samples equally, the existing contrastive learning framework is insufficient for learning discriminating user representations in recommendation. To bridge this research gap, we propose the Contrast over Contrastive Learning framework for training recommender models, named CCL4Rec, which models the nuances of different augmented views by further contrasting augmented positives/negatives with adaptive pulling/pushing strengths, i.e., the contrast over (vanilla) contrastive learning. To accommodate these contrasts, we devise the hardness-aware augmentations that track the importance of behaviors being replaced in the query user and the relatedness of substitutes, and thus determining the quality of augmented positives/negatives. The hardness-aware augmentation also permits controllable contrastive learning, leading to performance gains and robust training. In this way, CCL4Rec captures the nuances of historical behaviors for a given user, which explicitly shields off the learned user representation from the effects of noisy behaviors. We conduct extensive experiments on two micro-video recommendation benchmarks, which demonstrate that CCL4Rec with far less model parameters could achieve comparable performance to existing state-of-the-art method, and improve the training/inference speed by several orders of magnitude.