Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Autonomously Reconfigurable Artificial Neural Network on a Chip

Jin, Zhanpeng (2010) Autonomously Reconfigurable Artificial Neural Network on a Chip. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Primary Text

Download (4MB) | Preview


Artificial neural network (ANN), an established bio-inspired computing paradigm, has proved very effective in a variety of real-world problems and particularly useful for various emerging biomedical applications using specialized ANN hardware. Unfortunately, these ANN-based systems are increasingly vulnerable to both transient and permanent faults due to unrelenting advances in CMOS technology scaling, which sometimes can be catastrophic. The considerable resource and energy consumption and the lack of dynamic adaptability make conventional fault-tolerant techniques unsuitable for future portable medical solutions. Inspired by the self-healing and self-recovery mechanisms of human nervous system, this research seeks to address reliability issues of ANN-based hardware by proposing an Autonomously Reconfigurable Artificial Neural Network (ARANN) architectural framework. Leveraging the homogeneous structural characteristics of neural networks, ARANN is capable of adapting its structures and operations, both algorithmically and microarchitecturally, to react to unexpected neuron failures. Specifically, we propose three key techniques --- Distributed ANN, Decoupled Virtual-to-Physical Neuron Mapping, and Dual-Layer Synchronization --- to achieve cost-effective structural adaptation and ensure accurate system recovery. Moreover, an ARANN-enabled self-optimizing workflow is presented to adaptively explore a "Pareto-optimal" neural network structure for a given application, on the fly. Implemented and demonstrated on a Virtex-5 FPGA, ARANN can cover and adapt 93% chip area (neurons) with less than 1% chip overhead and O(n) reconfiguration latency. A detailed performance analysis has been completed based on various recovery scenarios.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Jin, Zhanpengzhj6@pitt.eduZHJ6
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairCheng, Allen
Committee MemberMickle, Marlin Hmickle@pitt.eduMICKLE
Committee MemberChang, Shi-Kuochang@cs.pitt.eduSCHANG
Committee MemberLevitan, Steven Plevitan@pitt.eduLEVITAN
Committee MemberJia, Wenyanwej6@pitt.eduWEJ6
Committee MemberMao, Zhi-Hongmaozh@engr.pitt.eduZHM4
Date: 30 September 2010
Date Type: Completion
Defense Date: 25 June 2010
Approval Date: 30 September 2010
Submission Date: 21 July 2010
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical Engineering
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: autonomous reconfiguration; bio-inspired; neuron virtualization; physical neurons; self-healing; self-optimizing; virtual neurons
Other ID:, etd-07212010-151144
Date Deposited: 10 Nov 2011 19:52
Last Modified: 15 Nov 2016 13:46


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item