IIT Guwahati researchers develop SH²RFSSM brain-inspired AI model for long sequences

Researchers at IIT Guwahati developed a brain-inspired AI model called “Spiking Heterogeneous Harmonic Resonate-and-Fire State Space Model” (SH²RFSSM) to process long sequential data with lower estimated energy consumption than many conventional approaches, according to the team. The researchers reported evaluating the model on 17 benchmark datasets and said it achieved performance comparable to state-of-the-art sequence models while using substantially less estimated energy.