Time-series foundation models show potential for E-Nose data with fine-tuning

A new paper explores the effectiveness of time-series foundation models (TSFMs) for electronic nose (E-Nose) data, a domain previously underexplored by these advanced models. The research assesses TSFMs like Chronos-2 and MOMENT, investigating their ability to generate useful embeddings for gas identification and concentration prediction. Findings indicate that while TSFMs show potential, fine-tuning is crucial for satisfactory performance on E-Nose data, and combining TSFM embeddings with specialized models can further boost accuracy. AI

IMPACT Suggests potential for advanced time-series models in gas-sensing applications, but highlights the need for domain-specific fine-tuning.

RANK_REASON Academic paper assessing existing models on a new domain. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Chronos 2 Forecasting Model
• Hugging Face
• MOMENT

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →