From prototype to training RLHF pipelines with TorchRL by Vincent Moens | Data Makers Fest 23

This presentation introduced TorchRL as a tool that enables rapid development of Reinforcement Learning from Human Feedback (RLHF) solutions. RLHF techniques are the foundation for numerous significant advancements in generative AI, such as chatbots and more. However, these techniques are still in their early stages, which makes it challenging to apply them directly to new domains or specific use cases.

TorchRL provided a modular and user-friendly API that offered various building blocks, granting users complete control over the RLHF training process. A hands-on demonstration showcased how these primitives enabled customized solutions for specific problems.

::::::

If you love watching talks like this one, consider joining us in person at the next event: www.datamakersfest.com

👉 FOLLOW US
Instagram: https://www.instagram.com/datamakersfest/
LinkedIn: https://www.linkedin.com/showcase/data-makers-fest/

Our channel features talks for anyone building products and services with and around data. Subscribe to our channel for videos on Data Science, Machine Learning, AI, Data Engineering, and more.

Data Makers Fest videos may be used for non-commercial purposes under a Creative Commons License, Attribution–Non-Commercial–No Derivatives (or the CC BY – NC – ND 4.0 International). To use the talk for other purposes, please contact us at hi@datamakersfest.com.

#datamakersfest #datascience #ai #machinelearning #dataengineering

Видео From prototype to training RLHF pipelines with TorchRL by Vincent Moens | Data Makers Fest 23 канала Data Makers Fest