CaReTS: A Multi-Task Framework Unifying Classification and Regression for Time Series Forecasting

2025-11-12

AI Summary

Existing deep learning models struggle to simultaneously achieve high accuracy and interpretability of temporal dynamics in multi-step time series forecasting. To address this, we propose CaReTS—a classification-and-regression collaborative multi-task framework that employs a dual-stream architecture: one stream models trend evolution via discrete classification, while the other captures numerical deviations through continuous regression. An uncertainty-aware adaptive multi-task loss dynamically balances task weights during training. CaReTS is encoder-agnostic, seamlessly integrating CNN, LSTM, and Transformer backbones to yield four variants (CaReTS1–4). Extensive experiments on multiple real-world datasets demonstrate that CaReTS consistently outperforms state-of-the-art methods, achieving significant improvements in both point prediction accuracy and trend identification accuracy. These results validate its effectiveness, generalizability across architectures and domains, and enhanced decision transparency through interpretable trend modeling.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Recent advances in deep forecasting models have achieved remarkable performance, yet most approaches still struggle to provide both accurate predictions and interpretable insights into temporal dynamics. This paper proposes CaReTS, a novel multi-task learning framework that combines classification and regression tasks for multi-step time series forecasting problems. The framework adopts a dual-stream architecture, where a classification branch learns the stepwise trend into the future, while a regression branch estimates the corresponding deviations from the latest observation of the target variable. The dual-stream design provides more interpretable predictions by disentangling macro-level trends from micro-level deviations in the target variable. To enable effective learning in output prediction, deviation estimation, and trend classification, we design a multi-task loss with uncertainty-aware weighting to adaptively balance the contribution of each task. Furthermore, four variants (CaReTS1--4) are instantiated under this framework to incorporate mainstream temporal modelling encoders, including convolutional neural networks (CNNs), long short-term memory networks (LSTMs), and Transformers. Experiments on real-world datasets demonstrate that CaReTS outperforms state-of-the-art (SOTA) algorithms in forecasting accuracy, while achieving higher trend classification performance.

Problem

• Unifying classification and regression for time series forecasting
• Disentangling macro trends from micro deviations in predictions
• Adaptively balancing multi-task learning with uncertainty weighting

Innovation

• Dual-stream architecture combining classification and regression
• Uncertainty-aware multi-task loss for adaptive learning
• Four variants incorporating CNN LSTM and Transformer encoders