Online Fine-Tuning of Carbon Emission Predictions using Real-Time Recurrent Learning for State Space Models

TL;DR

This paper presents a method for online fine-tuning of structured state space models to improve carbon emission predictions in real-time, especially useful in resource-limited, dynamic environments.

Contribution

It introduces a novel real-time recurrent learning approach for online adaptation of SSMs during inference, which was not previously possible.

Findings

Reduces prediction error during inference

Effective in resource-constrained environments

Demonstrates online adaptation benefits

Abstract

This paper introduces a new approach for fine-tuning the predictions of structured state space models (SSMs) at inference time using real-time recurrent learning. While SSMs are known for their efficiency and long-range modeling capabilities, they are typically trained offline and remain static during deployment. Our method enables online adaptation by continuously updating model parameters in response to incoming data. We evaluate our approach for linear-recurrent-unit SSMs using a small carbon emission dataset collected from embedded automotive hardware. Experimental results show that our method consistently reduces prediction error online during inference, demonstrating its potential for dynamic, resource-constrained environments.