Search Results for author: Cory Merkel

Found 9 papers, 0 papers with code

Model Extraction and Adversarial Attacks on Neural Networks using Switching Power Information

no code implementations15 Jun 2021 Tommy Li, Cory Merkel

Artificial neural networks (ANNs) have gained significant popularity in the last decade for solving narrow AI problems in domains such as healthcare, transportation, and defense.

Model extraction

On the Adversarial Robustness of Quantized Neural Networks

no code implementations1 May 2021 Micah Gorsline, James Smith, Cory Merkel

Reducing the size of neural network models is a critical step in moving AI from a cloud-centric to an edge-centric (i. e. on-device) compute paradigm.

Adversarial Robustness Model Compression +1

Exploring Energy-Accuracy Tradeoffs in AI Hardware

no code implementations17 Nov 2020 Cory Merkel

For simple binary decision problems with convolutional neural networks, it is shown that minimizing the cost corresponds to using fewer than the maximum number of resources (e. g. convolutional neural network layers and filters).

Decision Making

Energy Constraints Improve Liquid State Machine Performance

no code implementations8 Jun 2020 Andrew Fountain, Cory Merkel

A model of metabolic energy constraints is applied to a liquid state machine in order to analyze its effects on network performance.

Seizure Detection

An FPGA Implementation of a Time Delay Reservoir Using Stochastic Logic

no code implementations12 Sep 2018 Lisa Loomis, Nathan McDonald, Cory Merkel

This paper presents and demonstrates a stochastic logic time delay reservoir design in FPGA hardware.

General Classification

Current-mode Memristor Crossbars for Neuromemristive Systems

no code implementations17 Jul 2017 Cory Merkel

Motivated by advantages of current-mode design, this brief contribution explores the implementation of weight matrices in neuromemristive systems via current-mode memristor crossbar circuits.

Unsupervised Learning in Neuromemristive Systems

no code implementations27 Jan 2016 Cory Merkel, Dhireesha Kudithipudi

Neuromemristive systems (NMSs) currently represent the most promising platform to achieve energy efficient neuro-inspired computation.

Cannot find the paper you are looking for? You can Submit a new open access paper.