On Unlearning for Time-series Forecasting

2026-10-02

AI Summary

This study addresses the prohibitive retraining costs associated with data deletion in time series forecasting models and the instability of conventional machine unlearning methods. To this end, we propose RDTU, a novel framework that introduces the first retention-set-based Neural Tangent Kernel (NTK) basis prediction mechanism. By integrating diffusion models to generate residual-corrected pseudo-labels, RDTU effectively overcomes the parameter update instability induced by causal window coupling. Experimental results demonstrate that the unlearned models produced by RDTU achieve performance highly comparable to full retraining while significantly outperforming existing baselines. Ultimately, this work enables efficient, controllable time series data unlearning and lightweight model updates.

Abstract

Time-series forecasting is widely used in sensitive domains. Models in these settings are often trained on longitudinal user- or entity-level records, which may later require removal because they contain sensitive or proprietary information or have been corrupted by sensor failures. To address such deletion requests without costly retraining, machine unlearning has been widely studied as a practical mechanism for privacy protection and data governance. However, the application of machine unlearning to time series prediction has not yet been well realized; this is mainly due to the following unique challenges: Gradient-based unlearning can be unstable because a deleted observation participates in multiple causally connected forecasting windows, causing parameter updates to propagate beyond the requested interval and degrade retained forecasting utility. Label-guided updating offers a more controlled alternative, but continuous and context-dependent forecasts lack a suitable replacement target, while the exact-retrained output is unavailable during unlearning. Moreover, the remaining support for a deleted temporal pattern is highly non-uniform. Some affected windows retain structurally similar counterparts in the retained data, whereas others become underrepresented or isolated. We present RDTU, a Residual Diffusion framework for time-series unlearning. RDTU first uses a retained-set neural tangent kernel predictor to obtain a deletion-compatible base forecast. Then it quantifies the global and local structural support of each affected window using the volume contribution of the retained-reference data. Then a diffusion model generates a residual correction that estimates the counterfactual forecast, yielding a pseudo-label field that guides a lightweight model update. Experiments show that RDTU consistently produces unlearned models that most closely match exact retraining.

Problem

• Machine Unlearning
• Time-series Forecasting
• Privacy Protection
• Data Governance

Innovation

• Machine Unlearning
• Time-series Forecasting
• Residual Diffusion
• Neural Tangent Kernel
• Counterfactual Estimation