A Hardware Accelerator for Language-Guided Reinforcement Learning.
Aidin ShiriArnab Neelim MazumderBharat PrakashHouman HomayounNicholas R. WaytowichTinoosh MohseninPublished in: IEEE Des. Test (2022)
Keyphrases
- reinforcement learning
- low cost
- real time
- field programmable gate array
- programming language
- hardware and software
- state space
- language learning
- function approximation
- machine learning
- natural language
- massively parallel
- data mining
- specification language
- parallel implementation
- personal computer
- markov decision processes
- learning process
- supervised learning
- dynamic programming
- model free
- temporal difference
- object oriented programming
- multi agent
- hardware architecture
- policy search