Our CEO, Subutai Ahmad, will be delivering the keynote presentation at this year’s Efficient Generative AI Summit on September 11th in Santa Clara, CA. The Efficient Generative AI Summit focuses on successfully developing generative AI in a more affordable, sustainable and efficient manner by putting efficiency-boosting technologies in front of the vanguard of enterprises and software developers. This summit is also co-located with AI Hardware & Edge AI Summit, which takes place Sep 12-14 at the same venue.
Attend both shows to get a complete overview of software, systems, and hardware for fast and efficient AI training, deployment, and inference. Get a 20% discount by using the code NUMENTAPLATINUM20 at checkout. Register here.
Abstract:
Enterprises today face increasing pressure to incorporate GPT-scale large language models, or LLMs, throughout their product suites. However, these LLMs are complex, cumbersome, and costly to deploy. Unlike the human brain, which consumes a mere 30 watts of power, LLMs require massive compute and power footprints that are harming the planet. To address these issues, significant research is being conducted into LLM optimization but skyrocketing demand is outpacing these improvements. In this talk, we show how a new approach, grounded in neuroscience research, can significantly improve the performance of LLMs and provide a sustainable path forward.
In this presentation, Subutai will:
- Provide an overview of the Generative AI landscape and the challenges companies face in deploying LLMs.
- Review recent neuroscience discoveries that show how our brain achieves its incredible computational efficiency.
- Demonstrate how Numenta’s AI platform can apply these discoveries to AI and enable enterprises to deploy powerful LLMs at scale in an extremely cost-effective manner.
Subutai Ahmad • CEO