Rohan Varma

Abstract 4183: Finding Goldilocks: How AI-powered covalent drug discovery removes the “un” from “undruggable”

Johannes C. Hermann, Robert A. Everley, Han Wool Yoon, Rohan Varma et autres

Abstract Covalent drugs offer a path to drugging hard targets to provide urgently needed novel cancer medications. Starting from small covalent fragments is efficient in principle but difficult in practice due to challenges distinguishing generic from specific reactivity and the influence of …

us (code pays fourni par la source)

Training LLMs with Fault Tolerant HSDP on 100,000 GPUs

Omkar Salpekar, Rohan Varma, Kenny Yu, Vladimir Ivanov et autres

Large-scale training systems typically use synchronous training, requiring all GPUs to be healthy simultaneously. In our experience training on O(100K) GPUs, synchronous training results in a low efficiency due to frequent failures and long recovery time. To address this problem, we propose …

Training LLMs with Fault Tolerant HSDP on 100,000 GPUs

Omkar Salpekar, Rohan Varma, Kenny Yu, Vladimir Ivanov et autres

Large-scale training systems typically use synchronous training, requiring all GPUs to be healthy simultaneously. In our experience training on O(100K) GPUs, synchronous training results in a low efficiency due to frequent failures and long recovery time. To address this problem, we propose …

Learning a deep language model for microbiomes: The power of large scale unlabeled microbiome data

Quintin Pope, Rohan Varma, Christine Tataru, Maude M. David et autres

We use open source human gut microbiome data to learn a microbial "language" model by adapting techniques from Natural Language Processing (NLP). Our microbial "language" model is trained in a self-supervised fashion (i.e., without additional external labels) to capture the interactions among …

us (code pays fourni par la source)

Efficient Virtual Machine Placement Strategy Based on Enhanced Genetic Approach

Varun Barthwal, M. M. S. Rauthan, Rohan Varma, Sachin Gaur

in (code pays fourni par la source)

PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

Yanli Zhao, Andrew Gu, Rohan Varma, Liang Luo et autres

It is widely acknowledged that large models have the potential to deliver superior performance across a broad range of domains. Despite the remarkable progress made in the field of machine learning systems research, which has enabled the development and exploration of large …

Learning a deep language model for microbiomes: the power of large scale unlabeled microbiome data

Quintin Pope, Rohan Varma, Chritine Tataru, Maude M. David et autres

Abstract We use open source human gut microbiome data to learn a microbial “language” model by adapting techniques from Natural Language Processing (NLP). Our microbial “language” model is trained in a self-supervised fashion (i.e., without additional external labels) to capture the interactions …

us (code pays fourni par la source)

PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

Yanli Zhao, Andrew Gu, Rohan Varma, Liang Luo et autres

It is widely acknowledged that large models have the potential to deliver superior performance across a broad range of domains. Despite the remarkable progress made in the field of machine learning systems research, which has enabled the development and exploration of large …

SMA-LinR: An Energy and SLA-Aware Autonomous Management of Virtual Machines

Varun Barthwal, ManMohan Singh Rauthan, Rohan Varma

Cloud datacenters consume enormous energy and generate heat, which affects the environment. Hence, there must be proper management of resources in the datacenter for optimum usage of energy. Virtualization enabled computing improves the performance of the datacenters in terms of these parameters. …

in (code pays fourni par la source)

AntVMp: An Approach for Energy-Efficient Placement of Virtual Machines Using Max–Min Ant System

Varun Barthwal, Man Mohan Singh Rauthan, Rohan Varma

in (code pays fourni par la source)

MOOC Global Housing Design

Frederique van Andel, D.E. van Gameren, Nelson Mota, Rohan Varma

cn (code pays fourni par la source)

PyTorch distributed

Li Shen, Yanli Zhao, Rohan Varma, Omkar Salpekar et autres

This paper presents the design, implementation, and evaluation of the PyTorch distributed data parallel module. Py-Torch is a widely-adopted scientific computing package used in deep learning research and applications. Recent advances in deep learning argue for the value of large datasets and …

il, pl (code pays fourni par la source)