no code implementations • 15 Feb 2024 • Zain Taufique, Muhammad Awais Bin Altaf, Antonio Miele, Pasi Liljeberg, Anil Kanduri
We present a combinatorial evaluation of power-performance-accuracy trade-offs of EEG applications at different approximation, power, and performance levels to provide insights into the disciplined tuning of approximation in EEG applications on embedded platforms.
no code implementations • 16 Oct 2023 • Zain Taufique, Antonio Miele, Pasi Liljeberg, Anil Kanduri
Heterogeneity among edge devices and accuracy-performance trade-offs of DNN models present a complex exploration space while catering to the inference performance requirements.
no code implementations • 4 Aug 2022 • Anil Kanduri, Sina Shahhosseini, Emad Kasaeyan Naeini, Hamidreza Alikhani, Pasi Liljeberg, Nikil Dutt, Amir M. Rahmani
Smart eHealth applications deliver personalized and preventive digital healthcare services to clients through remote sensing, continuous monitoring, and data analytics.
no code implementations • 21 Feb 2022 • Sina Shahhosseini, Dongjoo Seo, Anil Kanduri, Tianyi Hu, Sung-soo Lim, Bryan Donyanavard, Amir M. Rahmani, Nikil Dutt
To this end, we propose a reinforcement-learning-based computation offloading solution that learns optimal offloading policy considering deep learning model selection techniques to minimize response time while providing sufficient accuracy.
no code implementations • 21 Feb 2022 • Sina Shahhosseini, Tianyi Hu, Dongjoo Seo, Anil Kanduri, Bryan Donyanavard, Amir M. Rahmani, Nikil Dutt
Furthermore, we deploy Hybrid Learning (HL) to accelerate the RL learning process and reduce the number of direct samplings.