Happy to share an optimized implementation of the 1D dilated convolution which helps us deliver scaling and industry-leading training-time performance for AtacWorks on Intel® Xeon® Cooper Lake, surpassing all published CPU-GPU results, : arxiv.org/abs/2104.08002

We are excited to team up with inspirational team at MILA to bring power of parallel computing to problems to have impact on humanity and enrich life of every person. #PopulationScaleAI Pradeep Dubey Intel AI

We did it again! 11 years back, PCL won the best paper award at SIGMOD This week we earned not just one, but two honors at SIGMOD: Bao wins Best Paper award, and its real-world application wins Honorable Mention awardl! MIT CSAIL lnkd.in/g2MaW3r lnkd.in/gqj7tyv

Honored to be chosen to keynote the computer architecture conference we architects have grown with and learnt from! Thank you ISCA! The Era of ubiquitous AI: A call-to-arms for Architects, June 16, 7am pacific iscaconf.org/isca2021/progr…

Thank you ISCA, Lizy John and Jose Martinez - for the keynote opportunity this morning! @profmartinez ISCA

Thank you Lizy for the video link of my keynote: youtube.com/watch?v=FeVa2q…

Inspiring to see India lead the world in democratizing AI like never before and challenging it to solve population-scale problems! And a proud moment to be contributing to this mission! Bharat Kaul Nivruti #iamintel

Looking forward to sharing with you in my Tech Insight session next week how we are making AI more scalable, productive, performant for your needs! #IntelOn

Excited about @habanalabs' Gaudi AI accelerators bringing a new level of #deeplearning and training efficiency to Amazon Web Services customers. Amazon EC2 DL1 instances will drive the next wave of #EmergingAI and enable more companies to implement #AI training. go.aws/3Cx6k9v

In case you missed my Tech Insight talk on AI at IntelON earlier this month or some of our progress updates at last week’s SC’21 … here is a summary writeup to get you up to speed: medium.com/intel-tech/ai-…

Excited to be part of @intelindia #allAI2021 summit on Dec 1st, a 24-hour virtual marathon event, bringing global leaders from govt, academia & industry: "Biology is eating the world & AI is eating Biology". Register now bit.ly/IntelAI21 #BeyondBorders

In case you missed my IntelON Tech Insight talk on "The Fast Path to Scale AI Everywhere", you can now watch it, along with other talks, sessions and the keynote: intel.com/content/www/us…

Happy to proud to share our lab's progress on speeding up end-to-end AI workflows! Sanchit Misra (I am hiring) Bharat Kaul

Celebrating three decades of Intel partnership with India!

We are ready to 'walk' our 'talk' aimed at shaping the future of biology and healthcare with AI. Join us if this excites you too!

We are happy to announce mat2qubit, a package for converting many quantum and classical problems into a representation that quantum computers can understand. It’s intended to be used for a variety of physics, materials, and classical optimization problems.

Happy and proud to share ... PolarFly: In a cross-geo collaboration with DoE Lab LANL, Colgate University and ETH Switzerland, we present first-known diameter-2 topology that asymptotically reaches the Moore bound. Also accepted at SC'22 arxiv.org/abs/2208.01695

Happy and proud to be part of this industry-wide initiative and specification for lower-precision FP8 arithmetic to make AI more performant, productive and energy-efficient: sgq.io/zx5Gkpa

DistGNN-MB: novel compute-communication overlap algorithm improves TTT for GraphSAGE on a 3rd gen Intel Xeon cluster by >5x over the widely used DistDGL. Stay tuned for optimization for our next-gen Xeon! arxiv.org/abs/2211.06385 Bharat Kaul Sanchit Misra (I am hiring)

Quantum getting ready for Deep Learning ... first experimental realization of our 2017 proposal (arXiv:1711.11240), thanks to our collaboration with DeCarlo's lab.