The National Science Foundation (NSF) awarded a $597,224 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) program to the University of California, Los Angeles (UCLA) to develop a novel class of feedforward neural networks that can process temporal information more effectively. The project, titled "RI: SMALL: ADAPTIVE SYNAPTIC DYNAMICS NEURAL NETWORKS: A NOVEL HYPOTHESIS FOR TEMPORAL COMPUTATIONS IN ARTIFICIAL NEURAL NETWORKS," aims to incorporate short-term synaptic plasticity into feedforward artificial neural networks to optimize the processing of time-varying information, such as speech. The research will contrast the performance of these adaptive synaptic dynamics networks against standard feedforward and recurrent neural networks on tasks like Morse code and speech recognition. No subawards are planned for this 3-year project, which is scheduled to run from October 2024 through September 2027.
Generated 3/4/25, 8:22 AM