Title :
A 345mW heterogeneous many-core processor with an intelligent inference engine for robust object recognition
Author :
Seungjin Lee ; Jinwook Oh ; Minsu Kim ; Junyoung Park ; Joonsoo Kwon ; Hoi-Jun Yoo
Author_Institution :
KAIST, Daejeon, South Korea
Abstract :
A 228GOPS 345 mW heterogeneous many-core processor combines bottom-up and top-down attention for high accuracy. It uses analog-digital mixed-mode neuro-fuzzy inference circuits to realize robust object recognition. Weight perturbation learning and workload-aware voltage/frequency control are adopted. The 5 × 10 mm2 chip with 96% recognition accuracy contains 4 SIMD vector processing elements and 32 MIMD processors.
Keywords :
frequency control; fuzzy reasoning; multiprocessing systems; neural nets; object recognition; perturbation techniques; voltage control; 32 MIMD processor; 4 SIMD vector processing element; analog-digital mixed mode neuro fuzzy inference circuits; heterogeneous many core processor; intelligent inference engine; power 3.5 mW; robust object recognition; weight perturbation learning; workload aware voltage-frequency control; Acceleration; Computer vision; Energy consumption; Engines; Forward error correction; Layout; Neurofeedback; Object recognition; Robustness; Throughput;
Conference_Titel :
Solid-State Circuits Conference Digest of Technical Papers (ISSCC), 2010 IEEE International
Conference_Location :
San Francisco, CA
Print_ISBN :
978-1-4244-6033-5
DOI :
10.1109/ISSCC.2010.5433905
