Analog dnn inference accuracy improves with temperature aware methods
Beyond Noise: Understanding and Overcoming Temperature Effects in Analog DNN Inference
Machine Learning
Summary
Analog computing devices can use less energy than regular digital ones, making them good for phones and embedded gadgets. But they can get less accurate when the temperature changes because parts behave differently. The authors studied how temperature affects these devices and found that consistent changes in components cause most problems, more than random noise. They tested different training and calibration methods and found that training with hardware and accounting for temperature help keep accuracy steady.
What this means in practice
- •For embedded system engineers: Maintain analog DNN accuracy across temperatures by integrating temperature-aware calibration in embedded device workflows.
- •For machine learning hardware developers: Use hardware-in-the-loop training techniques to robustly train analog accelerators against temperature-related accuracy drops.
Authors
Niklas Summ, Xiao Wang, Hendrik Borras, Bernhard Klein, Holger Fröning
Abstract
The energy efficiency of analog computing makes it one of the most promising candidates for deploying resource-intensive machine learning workloads on constrained platforms such as mobile and embedded devices. However, analog accelerators are inherently susceptible to noise and non-idealities arising from physical component variations, whose behavior is further sensitive to environmental factors. These effects can significantly degrade inference accuracy. In this work, we conduct a comprehensive experimental study on a representative example of analog hardware to investigate the impact of temperature. We first characterize the behavior of stochastic and systematic non-idealities across a range of operating temperatures. Following this, we compare a set of simulation-based and hardware-based mitigation strategies aimed at improving robustness against temperature-induced performance degradation. Our results suggest that temperature-induced degradation is driven primarily by systematic non-idealities rather than stochastic noise alone. Noise-aware training improves robustness, while hardware-in-the-loop training and temperature-aware calibration provide the strongest accuracy retention across varying thermal conditions.