Title | ||
---|---|---|
A 55-nm, 1.0–0.4V, 1.25-pJ/MAC Time-Domain Mixed-Signal Neuromorphic Accelerator With Stochastic Synapses for Reinforcement Learning in Autonomous Mobile Robots |
Abstract | ||
---|---|---|
Reinforcement learning (RL) is a bio-mimetic learning approach, where agents can learn about an environment by performing specific tasks without any human supervision. RL is inspired by behavioral psychology, where agents take actions to maximize a cumulative reward. In this paper, we present an RL neuromorphic accelerator capable of performing obstacle avoidance in a mobile robot at the edge of the cloud. We propose an energy-efficient time-domain mixed-signal (TD-MS) computational framework. In TD-MS computation, we demonstrate that the energy to compute is proportional to the importance of the computation. We leverage the unique properties of stochastic networks and recent advances in Q-learning in the proposed RL implementation. The 55-nm test chip implements RL using a three-layered fully connected neural network and consumes a peak power of 690
<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu \text{W}$ </tex-math></inline-formula>
. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/JSSC.2018.2881288 | IEEE Journal of Solid-State Circuits |
Keywords | Field | DocType |
Sensors,Neural networks,Hardware,Time-domain analysis,Neuromorphics,Mobile robots | Obstacle avoidance,Time domain,Computer science,Neuromorphic engineering,Electronic engineering,Artificial neural network,Computer engineering,Mobile robot,Reinforcement learning,Cloud computing,Computation | Journal |
Volume | Issue | ISSN |
54 | 1 | 0018-9200 |
Citations | PageRank | References |
2 | 0.38 | 0 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Anvesha Amaravati | 1 | 19 | 5.56 |
Saad Bin Nasir | 2 | 68 | 10.21 |
Justin Ting | 3 | 3 | 1.42 |
Insik Yoon | 4 | 9 | 3.36 |
Arijit Raychowdhury | 5 | 284 | 48.04 |