Publications

• Desh: Deep Learning for HPC System Health Resilience. Anwesha Das, Abhinav Vishnu, Charles Siegel and Frank Mueller. SC Poster, 2017

• What does fault tolerant Deep Learning need from MPI?. Vinay Amatya, Abhinav Vishnu, Charles Siegel and Jeff Daily. EuroMPI/USA, 2017.

• Generating Performance Models for Irregular Applications. Ryan Friese, Nathan Tallent, Abhinav Vishnu, Darren Kerbyson and Adolfy Hoisie. IPDPS, 2017.

• User-transparent Distributed TensorFlow. Abhinav Vishnu, Joseph Manzano, Charles Siegel and Jeff Daily. Arxiv Report, 2017.

• Comparing NVIDIA DGX-1/Pascal and Intel Knights Landing on Deep Learning Workloads. Nitin Gawande, Joshua Landwehr, Jeff Daily, Nathan Tallent, Abhinav Vishnu and Darren Kerbyson. ParLearning 2017

• Deep Learning for Computational Chemistry. Garrett Goh, Nathan Hodas and Abhinav Vishnu. Journal of Computational Chemistry (JCC'17)

• Adaptive Neuron Apoptosis for Accelerating Deep Learning on Large Scale Systems. Charles Siegel, Jeff Daily and Abhinav Vishnu. International Conference on Big Data (BigData'16)

• Fault Tolerant Frequent Pattern Mining. Sameh Shohdy, Abhinav Vishnu and Gagan Agrawal. International Conference on High Performance Computing (HiPC'16)

• Fault Tolerant Support Vector Machines. Sameh Shohdy, Abhinav Vishnu, and Gagan Agrawal. International Conference on Parallel Processing (ICPP'16)

• Accelerating Deep Learning with Shrinkage and Recall. Shuai Zheng, Abhinav Vishnu, and Chrish Ding. International Conference on Parallel and Distributed Systems (ICPADS'16).

• Distributed TensorFlow with MPI. Abhinav Vishnu, Charles Siegel and Jeff Daily. ArXiv Report, 2016

• Fault Modeling of Extreme Scale Applications using Machine Learning. Abhinav Vishnu, Hubertus van Dam, Nathan Tallent, Darren Kerbyson and Adolfy Hoisie. IEEE International Parallel and Distributed Processing Symposium (IPDPS), May, 2016

• Predicting the top and bottom ranks of billboard songs using Machine Learning. Vivek Datla and Abhinav Vishnu. ArXiv report, 2016.

• Fast and Accurate Support Vector Machines on Large Scale Systems. Abhinav Vishnu, Jeyanthi Narasimhan, and Lawrence Holder, Darren Kerbyson and Adolfy Hoisie. IEEE Cluster 2015, September, 2015

• Large Scale Frequent Pattern Mining using MPI One-Sided Model. Abhinav Vishnu, and Khushbu Agarwal. IEEE Cluster 2015, September, 2015

• Accelerating k-NN with Hybrid MPI and OpenSHMEM. Jian Lin, Khaled Hamidouche, Jie Zhang, Xiaoyi Lu, Abhinav Vishnu and Dhabaleswar Panda. OpenSHMEM Workshop, August, 2015.