no code implementations • 10 Mar 2022 • Shihao Song, Adarsha Balaji, Anup Das, Nagarajan Kandasamy
First, on the technology front, we propose an optimization scheme where the NVM resistance state that takes the longest time to sense is set on current paths having the least delay, and vice versa, reducing the average PE latency, which improves the QoS.
no code implementations • 27 Aug 2021 • Shihao Song, M. Lakshmi Varshika, Anup Das, Nagarajan Kandasamy
We propose an SDFG-based design flow for mapping spiking neural networks (SNNs) to many-core neuromorphic hardware with the objective of exploring the tradeoff between throughput and buffer size.
no code implementations • 4 Aug 2021 • Shihao Song, Harry Chong, Adarsha Balaji, Anup Das, James Shackleford, Nagarajan Kandasamy
We propose DFSynthesizer, an end-to-end framework for synthesizing SNN-based machine learning programs to neuromorphic hardware.
no code implementations • 5 May 2021 • Shihao Song, Jui Hanamshet, Adarsha Balaji, Anup Das, Jeffrey L. Krichmar, Nikil D. Dutt, Nagarajan Kandasamy, Francky Catthoor
We propose a new architectural technique to mitigate the aging-related reliability problems in neuromorphic systems, by designing an intelligent run-time manager (NCRTM), which dynamically destresses neuron and synapse circuits in response to the short-term aging in their CMOS transistors during the execution of machine learning workloads, with the objective of meeting a reliability target.
no code implementations • 4 May 2021 • Adarsha Balaji, Shihao Song, Twisha Titirsha, Anup Das, Jeffrey Krichmar, Nikil Dutt, James Shackleford, Nagarajan Kandasamy, Francky Catthoor
Recently, both industry and academia have proposed many different neuromorphic architectures to execute applications that are designed with Spiking Neural Network (SNN).
no code implementations • 9 Mar 2021 • Twisha Titirsha, Shihao Song, Anup Das, Jeffrey Krichmar, Nikil Dutt, Nagarajan Kandasamy, Francky Catthoor
We propose eSpine, a novel technique to improve lifetime by incorporating the endurance variation within each crossbar in mapping machine learning workloads, ensuring that synapses with higher activation are always implemented on memristors with higher endurance, and vice versa.
no code implementations • 19 Sep 2020 • Adarsha Balaji, Shihao Song, Anup Das, Jeffrey Krichmar, Nikil Dutt, James Shackleford, Nagarajan Kandasamy, Francky Catthoor
With growing model complexity, mapping Spiking Neural Network (SNN)-based applications to tile-based neuromorphic hardware is becoming increasingly challenging.
no code implementations • 10 Jun 2020 • Shihao Song, Anup Das, Nagarajan Kandasamy
We evaluate RENEU using different machine learning applications on a state-of-the-art neuromorphic hardware with NVM synapses.
1 code implementation • 7 Apr 2020 • Shihao Song, Adarsha Balaji, Anup Das, Nagarajan Kandasamy, James Shackleford
First, we propose a greedy technique to partition an SNN into clusters of neurons and synapses such that each cluster can fit on to the resources of a crossbar.
no code implementations • 1 Nov 2019 • Adarsha Balaji, Shihao Song, Anup Das, Nikil Dutt, Jeff Krichmar, Nagarajan Kandasamy, Francky Catthoor
Our framework first extracts the precise times at which a charge pump in the hardware is activated to support neural computations within a workload.
no code implementations • 14 Jun 2016 • Tingshan Huang, Harish Sethu, Nagarajan Kandasamy
This variance-based subspace approach to dimensionality reduction forces a fixed choice of the number of dimensions, is not responsive to real-time shifts in observed traffic patterns, and is vulnerable to normal traffic spoofing.