SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC On-Demand

AI Application Deployment and Inference
Presentation
Media
Protecting Pulsed High-Power Lasers with Real-Time Image Classification
Learn how to combine computer vision techniques and deep learning to improve the sensitivity of a real-time, GPU-powered safety system. In petawatt laser systems, firing at 10 Hz, suddenly appearing scatterers can damage components. Spreading of damage can be avoided by suspending operation immediately on occurrence of such an event. We'll present our approach for the automatic detection of critical failure states from intensity profiles of the laser beam. By incorporating quick feature detection and learned heuristics for feature classification, both real-time constraints and limited available training data are accommodated. Localization of triggering feature is crucial for when the problem is located in non-sensitive sections and will not be removed from the beam in production.
Learn how to combine computer vision techniques and deep learning to improve the sensitivity of a real-time, GPU-powered safety system. In petawatt laser systems, firing at 10 Hz, suddenly appearing scatterers can damage components. Spreading of damage can be avoided by suspending operation immediately on occurrence of such an event. We'll present our approach for the automatic detection of critical failure states from intensity profiles of the laser beam. By incorporating quick feature detection and learned heuristics for feature classification, both real-time constraints and limited available training data are accommodated. Localization of triggering feature is crucial for when the problem is located in non-sensitive sections and will not be removed from the beam in production.  Back
 
Keywords:
AI Application Deployment and Inference, Advanced AI Learning Techniques (incl. GANs and NTMs), Computer Vision, GTC Silicon Valley 2018 - ID S8330
Streaming:
Share:
Algorithms and Numerical Techniques
Presentation
Media
Non-Local Lattice Encoding for Bit-Vectorized Cellular Automata GPU Implementations
In many areas, from physics to economics to social sciences, there are problems that can be mapped to stochastic cellular automata (SCA). In combination with machine learning techniques, cellular automata with learned rules can be used to efficiently predict real-world systems. In physics, they are used to study atomistically the size and shape evolution of micro- and nanostructures, providing insights into processes of self-organization crucial to today's nanotechnology. We present an extremely efficient SCA implementation of a surface growth model using bit-vectorization enhanced by non-local encoding on GPU. The employed technique and non-local encoding can be transferred to other applications.
In many areas, from physics to economics to social sciences, there are problems that can be mapped to stochastic cellular automata (SCA). In combination with machine learning techniques, cellular automata with learned rules can be used to efficiently predict real-world systems. In physics, they are used to study atomistically the size and shape evolution of micro- and nanostructures, providing insights into processes of self-organization crucial to today's nanotechnology. We present an extremely efficient SCA implementation of a surface growth model using bit-vectorization enhanced by non-local encoding on GPU. The employed technique and non-local encoding can be transferred to other applications.  Back
 
Keywords:
Algorithms and Numerical Techniques, Computational Physics, GTC Silicon Valley 2016 - ID P6124
Download:
Share:
Artificial Intelligence and Deep Learning
Presentation
Media
Protecting Pulsed High-Power Lasers with Real-Time Object Detection

Learn how to train and employ state-of-the-art object localization in a real-time safety application. In Petawatt laser systems, firing at 10Hz, suddenly appearing scatterers can damage components. Damage(-spreading) can be avoided by suspending operation immediately upon occurrence of such an event.

We present our approach for the automatic detection of critical failure states from intensity profiles of the laser beam. In order to minimize the rate of false alarms, which would reduce productivity or even render our system useless, we refrain from general anomaly detection and instead detect known error patterns. In this talk we present how we fitted the You Look Only Once(YOLO) approach, which is suited to low-latency object detection, to our problem and how we adapted the required multi-step training protocol to the available experimental data.

Learn how to train and employ state-of-the-art object localization in a real-time safety application. In Petawatt laser systems, firing at 10Hz, suddenly appearing scatterers can damage components. Damage(-spreading) can be avoided by suspending operation immediately upon occurrence of such an event.

We present our approach for the automatic detection of critical failure states from intensity profiles of the laser beam. In order to minimize the rate of false alarms, which would reduce productivity or even render our system useless, we refrain from general anomaly detection and instead detect known error patterns. In this talk we present how we fitted the You Look Only Once(YOLO) approach, which is suited to low-latency object detection, to our problem and how we adapted the required multi-step training protocol to the available experimental data.

  Back
 
Keywords:
Artificial Intelligence and Deep Learning, GTC Europe 2018 - ID E8170
Streaming:
Download:
Share:
Computational Physics
Presentation
Media
Simulation of Surface Growth and Lattices Gases Using GPUs
Restricted solid on solid surface growth models can be mapped onto binary lattice gases. We show that efficient simulation algorithms can be realized on GPUs either by CUDA or by OpenCL programming. We consider a deposition/evaporation model following KardarParisiZhang growth in d+1 dimensions, related to the Asymmetric Simple Exclusion Process. Up to 100 - 400 x speedup can be achieved with respect to the serial code running on a I5 core. This permits studying disorder and aging behavior in these system.
Restricted solid on solid surface growth models can be mapped onto binary lattice gases. We show that efficient simulation algorithms can be realized on GPUs either by CUDA or by OpenCL programming. We consider a deposition/evaporation model following KardarParisiZhang growth in d+1 dimensions, related to the Asymmetric Simple Exclusion Process. Up to 100 - 400 x speedup can be achieved with respect to the serial code running on a I5 core. This permits studying disorder and aging behavior in these system.  Back
 
Keywords:
Computational Physics, HPC and Supercomputing, GTC Silicon Valley 2015 - ID P5259
Download:
Share:
 
Handling Domain Decomposition in Massively Parallel Implementations of Stochastic Lattice Models
Nanopatterning of surfaces and bulk materials is very important from molecular electronics to photovoltaics. But, in order to understand the underlying physics of self-organization, large scale atomistic simulations are crucial. Only stochastic models can bridge the gap from nano to micro, enabling simulations of micron-sized volumes, billions of atoms and study long-time evolution. Random site-selection is essential but can be harmed by domain decomposition in GPGPU. We present solutions by example of a dimer-model for KPZ surface growth.
Nanopatterning of surfaces and bulk materials is very important from molecular electronics to photovoltaics. But, in order to understand the underlying physics of self-organization, large scale atomistic simulations are crucial. Only stochastic models can bridge the gap from nano to micro, enabling simulations of micron-sized volumes, billions of atoms and study long-time evolution. Random site-selection is essential but can be harmed by domain decomposition in GPGPU. We present solutions by example of a dimer-model for KPZ surface growth.  Back
 
Keywords:
Computational Physics, Developer - Algorithms, GTC Silicon Valley 2015 - ID P5266
Download:
Share:
 
Efficient Correlation-Free Many-States Lattice Monte Carlo on GPUs
We'll present a method for highly efficient lattice Monte Carlo simulations with correlation-free updates. Achieving freedom from erroneous correlations requires random selection of lattice sites for updates, which must be restricted by suitable domain decomposition to create parallelism. While approaches based on caching limit the number of allowed states, the multisurface-type approach presented here allows arbitrarily complex states. The effectiveness of the method is illustrated in the fact that it allowed us to solve a long-standing dispute around surface growth under random kinetic deposition in the KPZ-universality class. The method has also been applied to Potts models and is suitable for spin-glass simulations, such as those required to test quantum annealers, like D-Wave.
We'll present a method for highly efficient lattice Monte Carlo simulations with correlation-free updates. Achieving freedom from erroneous correlations requires random selection of lattice sites for updates, which must be restricted by suitable domain decomposition to create parallelism. While approaches based on caching limit the number of allowed states, the multisurface-type approach presented here allows arbitrarily complex states. The effectiveness of the method is illustrated in the fact that it allowed us to solve a long-standing dispute around surface growth under random kinetic deposition in the KPZ-universality class. The method has also been applied to Potts models and is suitable for spin-glass simulations, such as those required to test quantum annealers, like D-Wave.  Back
 
Keywords:
Computational Physics, HPC and Supercomputing, GTC Silicon Valley 2017 - ID S7240
Download:
Share:
Molecular Dynamics
Presentation
Media
GPU-Based Atomistic Simulations on Spatio-Temporal Experimental Scales
The Kinetic Metropolis Lattice Monte-Carlo (KMC) method is a means of performing atomistic simulations of self-organization processes in solids at by far larger scales than those accessible via Molecular Dynamics. Employing a cellular automaton approach allows incorporation of many body interactions and external driving forces. Here, we present an efficient KMC implementation on single and multiple GPUs, which allows us to study phase separation and nanostructure-evolution at spatio-temporal experimental scales. The KMC implementation has been used to develop with industrial partners a new Si-based nanocomposite for next-generation thin-film solar cells.
The Kinetic Metropolis Lattice Monte-Carlo (KMC) method is a means of performing atomistic simulations of self-organization processes in solids at by far larger scales than those accessible via Molecular Dynamics. Employing a cellular automaton approach allows incorporation of many body interactions and external driving forces. Here, we present an efficient KMC implementation on single and multiple GPUs, which allows us to study phase separation and nanostructure-evolution at spatio-temporal experimental scales. The KMC implementation has been used to develop with industrial partners a new Si-based nanocomposite for next-generation thin-film solar cells.  Back
 
Keywords:
Molecular Dynamics, GTC Silicon Valley 2014 - ID P4154
Download:
Share:
Video and Image Processing
Presentation
Media
High-Bandwidth 3D Image Compression to Boost Predictive Life Sciences
Modern microscopes easily produce large data volumes (terabyte datasets) at high rate (1,000 megabytes/s is no exception) that makes using them almost impossible. Once an acquisition is started, it typically has to be stopped again as the hard drives run full. We'll share how GPUs helped us bring this nightmare to an end. We'll introduce our open-source package, called sqeazy, that is capable of compressing microscopic data at faster speeds than a hard drive can spin. We show how GPUs provided a crucial boost in this endeavor and we'll share what technical challenges we overcame interfacing with modern video encoding libraries, like libavcodec of ffmpeg. Finally, we'll discuss how NVENC provides portable performance that helps scientists to observe living developing specimens over long time spans. This may be the foundation for modern predictive biology of the 21st century. Join us for a tour on how modern media technology straight from Hollywood can boost science!
Modern microscopes easily produce large data volumes (terabyte datasets) at high rate (1,000 megabytes/s is no exception) that makes using them almost impossible. Once an acquisition is started, it typically has to be stopped again as the hard drives run full. We'll share how GPUs helped us bring this nightmare to an end. We'll introduce our open-source package, called sqeazy, that is capable of compressing microscopic data at faster speeds than a hard drive can spin. We show how GPUs provided a crucial boost in this endeavor and we'll share what technical challenges we overcame interfacing with modern video encoding libraries, like libavcodec of ffmpeg. Finally, we'll discuss how NVENC provides portable performance that helps scientists to observe living developing specimens over long time spans. This may be the foundation for modern predictive biology of the 21st century. Join us for a tour on how modern media technology straight from Hollywood can boost science!  Back
 
Keywords:
Video and Image Processing, Healthcare and Life Sciences, GTC Silicon Valley 2017 - ID S7247
Download:
Share: