Srinath Obla

Srinath Obla

Auto-generated profile, based on publicly available information

Srinath Obla

Department of Computer Science, Rochester Institute of Technology, Rochester, NY 14623 USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 August 2020 in IEEE Access
Verified
Journal Article
Open access
SOSrinath Obla
Srinath Obla
XGXinghan Gong
Xinghan Gong
AAAsma Aloufi
Asma Aloufi
PHPeizhao Hu
Peizhao Hu
DTDaniel Takabi
Daniel Takabi
CryptoNets and subsequent work have demonstrated the capability of homomorphic encryption (HE) in the applications of private artificial intelligence (AI). In convolutional neural networks (CNNs), many computations are linear functions such as the convolution layer which can be homomorphically evaluated. However, there are layers such as the activation layer which is comprised of non-linear functions that cannot be homomorphically evaluated. One of the most commonly used methods is approximating these non-linear functions using low-degree polynomials. However, using the approximated polynomials as activation functions introduces errors which could have a significant impact on accuracy in classification tasks. In this paper, we present a systematic method to construct HE-friendly activation functions for CNNs. We first determine what properties in a good activation function contribute to performance by analyzing commonly used functions such as Rectified Linear Units (ReLU) and Sigmoid. Then, we compare polynomial approximation methods and search for an optimal range of approximation for the polynomial activation. We also propose a novel weighted polynomial approximation method tailored to the output distribution of a batch normalization layer. Finally, we demonstrate the effectiveness of our method using several datasets such as MNIST, FMNIST, CIFAR-10.