SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC On-Demand

Deep Learning and AI
Presentation
Media
Learning Particle Physics by Example: Using Generative Adversarial Networks to Accelerate Physics
Luke deOliveira (Berkeley Lab)
We'll discuss the use of generative modeling, in particular, generative adversarial networks (GANs), as a potential tool for speeding up expensive theoretical models and simulations in scientific and engineering applications. In particular, we'll s ...Read More
We'll discuss the use of generative modeling, in particular, generative adversarial networks (GANs), as a potential tool for speeding up expensive theoretical models and simulations in scientific and engineering applications. In particular, we'll show that GANs can learn high-energy particle physics by example - that is, we're able to generate physically meaningful processes from an end-to-end trained deep generative network. Our model presents a speedup of 200x using GPUs over the fastest current particle generation software. We'll walk through architectural choices given a very sparse domain, introduce a new dataset for benchmarking generative models, and discuss model evaluation for generative models in high-energy particle physics. Finally, we'll discuss future work in both high-energy particle physics and beyond.  Back
 
Keywords:
Deep Learning and AI, Computer Vision and Machine Vision, GTC Silicon Valley 2017 - ID S7666
Download: