Future-Conditioned Recommendations with Multi-Objective Controllable Decision Transformer

TL;DR

This paper introduces MocDT, a novel offline RL model that enables future-conditioned, multi-objective recommendations by directly specifying future goals, improving control and adaptability in recommender systems.

Contribution

The paper proposes a future-conditioned, multi-objective recommendation approach using MocDT, a new offline RL model that maps multiple objectives to item sequences for better control.

Findings

MocDT effectively generates item sequences aligned with specified objectives.

The model maintains competitive performance across various objectives.

It demonstrates improved flexibility in multi-objective recommendation scenarios.

Abstract

Securing long-term success is the ultimate aim of recommender systems, demanding strategies capable of foreseeing and shaping the impact of decisions on future user satisfaction. Current recommendation strategies grapple with two significant hurdles. Firstly, the future impacts of recommendation decisions remain obscured, rendering it impractical to evaluate them through direct optimization of immediate metrics. Secondly, conflicts often emerge between multiple objectives, like enhancing accuracy versus exploring diverse recommendations. Existing strategies, trapped in a "training, evaluation, and retraining" loop, grow more labor-intensive as objectives evolve. To address these challenges, we introduce a future-conditioned strategy for multi-objective controllable recommendations, allowing for the direct specification of future objectives and empowering the model to generate item sequences that align with these goals autoregressively. We present the Multi-Objective Controllable Decision Transformer (MocDT), an offline Reinforcement Learning (RL) model capable of autonomously learning the mapping from multiple objectives to item sequences, leveraging extensive offline data. Consequently, it can produce recommendations tailored to any specified objectives during the inference stage. Our empirical findings emphasize the controllable recommendation strategy's ability to produce item sequences according to different objectives while maintaining performance that is competitive with current recommendation strategies across various objectives.