Train host and infer reasoning models on Microsoft Foundry | BRK210 скачать в хорошем качестве

Train host and infer reasoning models on Microsoft Foundry | BRK210

Step through training and deploying a custom reasoning model with Azure ML and Microsoft Foundry. See the magic happen in real time and learn how to bring it to your own projects with Foundry. To learn more, please check out these resources: https://aka.ms/ignite25-plans-BestMod... 𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀: Vijay Aski Chris Lauren Ankit Parag Shah Wei Wei 𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻: This is one of many sessions from the Microsoft Ignite 2025 event. View even more sessions on-demand and learn about Microsoft Ignite at https://ignite.microsoft.com BRK210 | English (US) | Innovate with Azure AI apps and agents, Microsoft Foundry Breakout | Intermediate (200) #MSIgnite, #InnovatewithAzureAIappsandagents Chapters: 0:00 - Customers Using AI Foundry for Real-World ROI 00:02:09 - Challenges in Moving AI from Prototype to Production 00:15:29 - Reference to Research Paper on Harnessing LLMs for Business Insights 00:17:26 - Demonstration of Financial Analyst Agent Using Foundry 00:21:40 - Introduction of Reinforcement Learning for Accuracy Improvement 00:25:37 - Comparison of RFT generalization and multitask capabilities 00:28:39 - Distinction between synchronous and asynchronous reinforcement learning techniques 00:31:32 - Evaluation pipeline comparing different fine-tuning methods and performance results 00:42:54 - Benchmarking results show improved throughput and token latency