COMET-SG1: Lightweight Autoregressive Regressor for Edge and Embedded AI
This addresses the problem of stable, low-resource prediction for edge and embedded systems, though it appears incremental as it builds on existing regression and lightweight model concepts.
The paper tackles time-series prediction for edge and embedded AI by introducing COMET-SG1, a lightweight autoregressive regressor that achieves competitive short-horizon accuracy and significantly reduces long-horizon drift compared to baselines like LSTM and MLP.
COMET-SG1 is a lightweight, stability-oriented autoregressive regression model designed for time-series prediction on edge and embedded AI systems. Unlike recurrent neural networks or transformer-based sequence models, COMET-SG1 operates through linear behavior-space encoding, memory-anchored transition estimation, and deterministic state updates. This structure prioritizes bounded long-horizon behavior under fully autoregressive inference, a critical requirement for edge deployment where prediction errors accumulate over time. Experiments on non-stationary synthetic time-series data demonstrate that COMET-SG1 achieves competitive short-horizon accuracy while exhibiting significantly reduced long-horizon drift compared to MLP, LSTM, and k-nearest neighbor baselines. With a compact parameter footprint and operations compatible with fixed-point arithmetic, COMET-SG1 provides a practical and interpretable approach for stable autoregressive prediction in edge and embedded AI applications.