GDC + VRDC 2019 Session Scheduler

View, browse and sort the ever-growing list of GDC sessions by pass type, track, format, and more! With Session Scheduler, create an account to build your own GDC schedule and access it anywhere, including on the GDC app, once live.

If you've registered for GDC, you can use the same login credentials. Adding a session to your schedule does not guarantee you a seat. Sessions do fill up, so please arrive early to sessions that you would like to attend.

Introducing DirectML: How to Move Machine Learning from Research into Production (Presented by Microsoft)

Kevin Cogger (Sr. Software Engineer, Team Xbox, Microsoft)

Adrian Tsai (Sr. Software Engineer, Silicon Graphics & Media, Microsoft)

Location: Room 2009, West Hall

Date: Thursday, March 21

Time: 4:00pm - 5:00pm

Pass Type: All Access, GDC Conference + Summits, GDC Conference, GDC Summits, Expo Plus, Audio Conference + Tutorial, Indie Games Summit - Get your pass now!

Topic: Programming

Format: Sponsored Session

Vault Recording: Not Recorded

Audience Level: Advanced

**GDC Announcement**
Innovation with machine learning in gaming creates a demand for real-time inferencing, capable of integrating well with high performance applications. Do you have a machine learning model? Do you need machine learning for your rendering workloads, tooling, game engine or content pipeline?

This 401-level session will cover DirectML, a fast and powerful DirectX library that meets game developers' needs by providing hardware-accelerated operators to take your neural networks from research to production. For this demo we will show how to achieve super-resolution with a DirectML engine plug-in and bring our collaborators on stage to walk through examples of using DirectML to power their machine learning workloads.

Takeaway

- This talk is about giving the developers already working on ML, to get the tools they need to move into production.
- DirectML unlocks the power of the GPU for developers, with one recent example with a graphics application.

Intended Audience

- Developers already working in ML with a working learning model.
- Developers interested in looking ahead a couple of steps for how Microsoft and its partners are making ML work in games.