Speaker
Maryam Eslami
(Ruhr-Universität Bochum)
Description
Deep neural networks (DNNs) have been applied across diverse domains, including safety-critical applications. Past studies indicate that DNNs are very sensitive to changes in weights and activations due to uneven bit-weight distribution in standard number formats like fixed points, which can cause significant output accuracy fluctuations. To address this issue, we introduce a new data type called MONO to enhance bit flip resilience using uniformity at the bit level by employing symmetric weights for all bit positions. On average, MONO has improved error resilience more effectively than the fixed-point data type, even when utilizing triple modular redundancy (TMR) and most significant bit (MSB) protection, while maintaining low overhead.
| Talk's Q&A | During the talk |
|---|---|
| Talk duration | 15'+7' |
| Will you be able to present in person? | No |
| If we are unable to accommodate your oral presentation, would you be willing to consider presenting a poster instead? | Yes |
Author
Maryam Eslami
(Ruhr-Universität Bochum)
Co-authors
Prof.
Akash Kumar
Dr
Salim Ullah
Mr
Yuhao Liu