Browse all available sessions and instructor-led trainings below. Click “Add to Schedule” to build your personal agenda and reserve your place at instructor-led trainings.
Please note:
To attend instructor-led trainings, you must have a Conference & Training pass. You must also reserve your place at each instructor-led training you wish to attend. If you have reserved a place at an instructor-led training, please arrive on time to ensure you do not lose your seat.
Attendance to sessions is first come, first served. Please arrive early to the sessions you wish to attend to guarantee entry.
11.15 - #AI4Good on PowerAI: How can your coding skills help others? (Hackathon) Speaker: Carmen Recio
12:00 - Watson-Studio: Putting AI to Work for Business: Umit Mert Cakmak
13:00 - Virtualization is real: Cloud-enabled CAD (Cloud session) Speaker:Alex Hudak, Charlie Dawson (IMSCAD)
14:00 - Integrated IBM AI session (Systems & Cloud) Speaker: Florin Manaila, Alex Hudak
15:00 - Fuel Pipeline for AI (Storage Session),Daniel Reiberg
16:00 - Accelerate and Scale High Performance Computing with IBM Cloud and Rescale, Speaker: Jerry Gutierrez, IBM & Joris Poort, Rescale
5 Hours
Talk
Dilek Sezgün - AI, Big Data and Open Source Leader for Systems and Ecosystems, IBM Carmen Recio - Cognitive systems technical team and IBM Quantum Ambassador, IBM Joris Poort - Co-founder and CEO, Rescale Florin Manaila - Cognitive Systems Architect, IBM Sumit Gupta - VP, HPC, AI & Machine Learning, IBM Daniel Reiberg - Architect - Cognitive Systems, IBM Jerry Gutierrez - World Wide HPC Solution Leader, IBM Alex Hudak - IBM Cloud GPU Offering Manager, IBM
Join us for this event to hear from NVIDIA, key experts, technology leaders, researchers, and engineer on the latest innovations and solutions. See how they’re pushing the limits of technology to create the next generation of autonomous machines--from robotics and research to aerospace and disaster relief.
5 Hours
Workshop
Muralimohan Gopalakrishna, NVIDIA Bill Chou - Product Manager, MathWorks Ram Kokku - Development Manager, MathWorks Patrick Dietrich - CTO, Connect Tech Inc. Michael GIELDA - VP Business Development, Antmicro Gary Hilgemann - CTO, raypack.ai Norbert Kuperjans - Business Development, PC Partner Evan Chen - Chairman, Appro Photoelectron Inc. Michael Suppa - CEO, Roboception GmbH ZANE TSAI - Director, ADLINK Technology Amir Nahvi - Head of Hardware, Advertima AG
E8497 -
Women At GTC Lunch: "Leadership In The Age of AI"
This session will feature at the Women at GTC Lunch event and will explore the topic of "Leadership In The Age of AI". All attendees who support diversity and inclusion in the tech world are welcome to join.
120-Minutes
Panel
Polina Mamoshina - Senior Research Scientist, Insilico Medicine, Inc Vidya Munde Müller - Social Tech Entrepreneur, Women in AI Ambassador, Germany Silja Pieh - CFO & Head of Product Management, AID - Autonomous Intelligent Driving GmbH
Data is the lifeblood of the modern city, AI is the key to turning this information into insight. Deep Stream & NVIDIA AI solutions help capture, inspect and analyze data, to benefit everything from traffic and parking management to law enforcement and city services. AGENDA: We'll be addressing the biggest challenges in developing Intelligent Video Analytics (IVA) applications at scale and how NVIDIA SDKs and advances in AI can help solve them, quickly building solutions to process images and video for applications such as public safety, retail, traffic and transportation. We'll also offer a glimpse into what's to come in the future
Image (or semantic) segmentation is the task of placing each pixel of an image into a specific class. Learn how to segment MRI images to measure parts of the heart by:
· Comparing image segmentation with other computer vision problems
· Experimenting with TensorFlow tools such as TensorBoard and the TensorFlow Python API
· Learning to implement effective metrics for assessing model performance
Upon completion, you’ll be able to set up most computer vision workflows using deep learning.
Prerequisites: Basic experience with neural networks
90 minutes
DLI Instructor-Led Training
Adolf Hohl - Solution Architect, NVIDIA
Get started with an introduction to object detection and image segmentation. You'll explore the shift from traditional computer vision techniques to innovative methods based on deep learning and convolution neural networks (CNNs) for object detection. You'll learn how to:
· Leverage CNNs to iteratively improve performance and expand image understanding capabilities
· Use Microsoft Common Object in Contexts dataset and the Google object detection API in TensorFlow
· Determine accuracy vs. performance trade-offs between Single Shot Multibox Detectors (SSDs), Faster R-CNN with residual networks, and Mask R-CNN
Upon completion, you'll understand how to implement object detection networks with the API in TensorFlow.
Prerequisites: None
90 minutes
DLI Instructor-Led Training
Jan Jamaszyk - Solution Architect, NVIDIA
Anomaly detection is critical in many industries, especially cybersecurity, finance, healthcare, retail and telecom. Variational autoencoders can outperform traditional techniques for anomaly detection. In this mini course, you'll learn how to:
· Use Bayesian inference roots of variational autoencoders and their implementation
· Define anomalies as the probability of being generated from a given model below a certain threshold and set thresholds that are specific to industry or use case
· Use variational autoencoders to detect anomalies from all data points
Upon completion, you'll know how to train a variational autoencoder to detect anomalies within the data.
Prerequisites: Experience with CNNs
90 minutes
DLI Instructor-Led Training
Eric Harper - Deep Learning Solutions Architect, NVIDIA
E8108 -
GPU-Accelerated Computing at Scale
In pushing the limits of throughput of floating-point operations, GPUs have become a unique technology. During this session, we'll explore the current state of affairs from an application perspective. For this, we'll consider different computational science areas including fundamental research on matter, materials science, and brain research. Focusing on key application performance characteristics, we review current architectural and technology trends to derive an outlook towards future GPU-accelerated architectures.
45-minute
Talk
Dirk Pleiter - Group Leader, Forschungszentrum Jülich
E8231 -
The Big Trends in AI and How They are Affecting Companies
Explore the worldwide trend of implementing AI and Deep Learning solutions across a range of vertical markets as well as how they are a game-changing factor for the workflows of a company.
20 Minutes
Talk
Andreas Liebl - Managing director, Applied AI Andreas Braun - Managing Director, Head of Data & Applied AI, Europe, Accenture
E8300 -
NVIDIA's VR Insights: OpenGL, Vulkan and Dual-Input HMDs
This talk will feature an update on what's happening in the professional VR space at NVIDIA. We first introduce OpenGL and Vulkan VR functionality, and then will talk about how to drive dual-input HMDs from two GPUs efficiently.
45-minute
Talk
Ingo Esser - Sr. DevTech Engineer, NVIDIA Robert Menzel - Developer Technology Engineer, NVIDIA
E8311 -
GPU-Driven Deep Learning Paves New Ways for Drug Discovery Through High-Content Imaging
Drug discovery is focused around finding relationships between chemical structure and biological effects of small molecules. Since such models depend on already-investigated chemical structures, they can hardly propose completely novel chemical scaffolds, a problem which has always been a drawback of traditional drug design. Now, researchers from Johannes Kepler University Linz, together with Janssen Pharmaceuticals, have found a novel way to discover drugs with GPU-based Deep Learning: instead of the chemical structure, they use images of cells that were treated with small molecules, and leverage deep neural networks to find relationships with biological effects. Thus, the image-based strategy can propose completely new chemical scaffolds, since there is no dependency on known, well-investigated chemical structures. In ongoing drug discovery projects, this novel GPU-driven strategy has identified many novel chemical scaffolds and thereby increased the discovery rate of drug candidates by 60 and 250-fold.
45-minute
Talk
Sepp Hochreiter - Professor, Head of Institute, Johannes Kepler University Günter Klambauer - PhD, Johannes Kepler University
E8363 -
Machine Learning Techniques for Track Reconstruction at the CMS Experiment
Starting from 2020, the increased accelerator luminosity of the Large Hadron Collider at CERN will directly result in an increased number of simultaneous proton-proton collisions (pile-up) which will pose significant new challenges for the CMS experiment. The adoption of machine learning techniques, along with traditional algorithms implemented on GPUs for both training and online inference, would reduce the incremented workload for the track reconstruction and improve the event selection.
45-minute
Talk
Maurizio Pierini - Research Staff, CERN
E8399 -
The Reinvention of the Car – and How to Consider Safety Aspects for Launching Autonomous Driving
Clearly, Autonomous Driving has the unique potential to change the way we think about transportation. The rapid evolution of sensors, artificial intelligence and IT-infrastructure paves the way to a driverless future much faster than many think.
Let's have a look at what is out on the streets today and how we approach the fascinating future of Autonomous Driving at Mercedes-Benz.
Title:
The Reinvention of the Car - How to Consider Safety Aspects for Launching Autonomous Driving
Abstract:
Clearly, Autonomous Driving has the unique potential to change the way we think about transportation. The rapid evolution of sensors, artificial intelligence and IT-infrastructure paves the way to a driverless future much faster than many think.
Let's have a look at what is out on the streets today and how we approach the fascinating future of Autonomous Driving at Mercedes-Benz.
Starting with short Video CASE @ Daimler (Connected, Autonomous, Shared, Electric)
1. Innovations in series cars: ADAS Update
• History of ADAS Systems
• Current Level of Automation (Driver Assistance Systems)
• Field Validation
2. Why full vehicle automation makes sense: Motives
• Reasons for Vehicle Automation
3. Where are we heading to: Technology
• Sensor Setup
• Use Case
• How to understand Sensor Data
• How to collect the needed information
• Short Demo Drive (Video)
• Experience counts (field testing around the world)
• Safe System Architecture
• ISO 26262 Development for AD
• Strong Partners
• Outlook
45-minute
Talk
Bernhard Weidemann - Product and Technology Communications, Daimler AG Michael Hafner - Head of Automated Driving, Daimler
E8402 -
Augmented Material Creation using Substance at BMW Design Group
BMW Design Visualisation will demonstrate live how Allegorithmic's Substance Software is being used to create proper material and texture's content combining procedural and scan based information for their hyper-realistic real-time VR experiences. Substance PBR material description format has allowed them to leverage a unique library of materials across all their VR tools and workflow. We'll also cover car staging within an adapted environment, a key step for better VR immersion.
45-minute
Talk
Max Bostock - Visualization and VR Specialist, BMW Design Group Pierre Maheut - Market Strategy Director for Architecture and Design, Allegorithmic
As many have argued, the rise of artificial intelligence could be seen as the second machine age. The exponential growth in computing power, memory capacity, distributional modules, and most importantly the data (structured and unstructured) created, all require algorithmic innovations to dig out meaningful business insights. Machine learning, especially deep learning, is undoubtedly a crucial key to future. The latest development in computer vision, NLU, and virtual assistants, could easily improve / alter our current insurance structures. Generative models have the potential to reveal deep insights of clients, which hence lead to deep underwriting / customized healthcare strategy.
45-minute
Talk
Yuanyuan Liu - Director, Machine Learning, AIG
E8490 -
Path to future AI, scenarios & laws: Presented by HPE
AI has proceeded over the last 50 years in fits and starts and today the momentum has once again picked up dramatically. New discoveries and techniques in Deep Learning, the best performing AI, are emerging at a rate of one every two months. If anything, this pace is likely to accelerate in the future as this becomes relevant across many industries. While we are all worried about how to faster accelerate our own workloads, we need to also start thinking about what will come in 10 or 50 years' time. Will we be building an AI beneficial for all or for just some? Are we sure that we are putting the right goals into our AI systems? Shouldn't we make sure that the goals of a General AI are aligned with ours? What are the possible scenarios looking ahead? Let's take an objective look and discuss.
45-minute
Talk
Sorin Cheran - Technology Strategist, PhD, HPE
Join our Data Science Workshop, where industry leadership and data science luminaries will come together for an open dialogue discussing GPU-accelerated data science. Learn how NVIDIA and the ecosystem are advancing data science using open-source strategies to enable more innovative solutions.
AGENDA:
1:30 – 1:45 pm Introduction - Clément Farabet, VP of AI Infrastructure, NVIDIA
1:45 – 2:30 pm RAPIDS: The Platform Inside and Out – Josh Patterson, Director of AI Infrastructure, NVIDIA
2:30 – 3:15 pm Open Source Data Science and Where It’s Going: Panel
Panelists:
Wes McKinney, Managing Director of Ursa Labs
Travis Oliphant, Founder and CEO of Quansight
Peter Wang, CTO Anaconda
2 Hour
Workshop
Clement Farabet - VP, AI Infrastructure, NVIDIA Joshua Patterson - Director of AI Infrastructure, NVIDIA Wes McKinney - Director, Ursa Labs Travis Oliphant - CEO, QUANSIGHT Peter Wang, Anaconda Powered By Continuum Analytics
Designing an autonomous machine are about much more than just the AI. Electrical, Mechanical, Connectivity, and Security are just a few of the disciplines where you will require expertise. Not all companies will have complete expertise in all these areas. In this session, we will provide examples followed by design considerations, strategies and solutions to begin to address these challenges.
Explore an introduction to Numba, a just-in-time function compiler that allows developers to utilize the CUDA platform in their Python applications. You'll learn how to:
· Decorate Python functions to be compiled by Numba
· Use Numba to GPU accelerate NumPy ufuncs
Upon completion, you'll be able to use Numba to GPU-accelerate NumPy ufuncs in your Python code, and will be ready to learn how to write custom CUDA kernels in Python.
Prerequisites: Basic Python and Numpy Competency
90 minutes
DLI Instructor-Led Training
Issam Said - Solution Architect, NVIDIA
E8110 -
Deep Learning Demystified
See how deep neural networks are trained to perform tasks with super-human accuracy and will explore which deep neural network models are best-suited for a variety of tasks.
45-minute
Talk
Will Ramey - Sr. Director Global Head of Developer Programs, NVIDIA
E8150 -
The Journey from a Small Development Lab Environment to a Production GPU Inference Datacenter
We'll do a dive deep into best practices and real world examples of leveraging the power and flexibility of local GPU workstations, such as the DGX Station, to rapidly develop and prototype deep learning applications. This journey will take you from experimenting and iterating fast and often, to obtaining a trained model, to eventually deploying scale-out GPU inference servers in a datacenter. Tools available, that will be explained, are NGC (NVIDIA GPU Cloud), TensorRT, TensorRT Inference Server, and YAIS.
45-minute
Talk
Ryan Olson - Solutions Architect, NVIDIA Markus Weber - Senior Product Manager, NVIDIA
E8195 -
Accelerating Weather Prediction with OpenACC
In this talk attendees will learn how key algorithms for Numerical Weather Prediction were ported to the latest GPU technology and the substantial benefits gained from doing so. We will showcase the power of individual Voltas and the impressive performance of the cutting edge DGX-2 server with multiple GPUS connected by a high speed interconnect.
45-minute
Talk
Paddy Gillies - Code efficiency analyst, ECMWF Alan Gray - CUDA Dev Tech, NVIDIA
E8212 -
Attacking Financial Statement Audits with Adversarial Accounting Records
This session will explore how auditors can be misguided or "fooled" by adversarial accounting records or adversarial financial transactions.
Recent discoveries in deep learning research revealed that learned models are vulnerable to "adversarial examples," or a sample of slightly modified input data that intends to cause a human and/or machine to misclassify it. Such examples exhibit the potential to be dangerous, since they could be specifically designed to misguide auditors or an accountant. Securing accounting information systems against such "attacks" can be difficult. In this talk, we'll explain why such "adversarial examples" are of vital relevance in the context of fraud detection and financial statement audits. We will demonstrate how autoencoder neural networks can be trained in an adversarial setup to generate "fake" accounting records or financial transactions. Such financial transactions might be misused to "attack" an organization's internal control system or obfuscate fraudulent activities. The training of such examples was conducted by training several adversarial autoencoders using NVIDIA's DGX-1 system.
45-minute
Talk
Marco Schreyer - Deep Learning Researcher, German Research Center for AI (DFKI) Timur Sattarov - Forensic Data Analyst, PricewaterhouseCoopers GmbH WPG
E8261 -
Feature Visualisation Techniques for Medical Image Analysis
Learn how feature visualization techniques can be used for deep-learning-based medical image analysis. This talk features an introduction to feature visualization techniques and their practical application to the domain of medical image understanding and computer aided diagnosis. Learn about our latest research on obtaining high quality visualizations for recent network architectures and obtain guidelines for using them for your application. After attending this session, you will be able to answer the following questions: What are the implementation caveats of feature visualization techniques? Which conclusions may I draw from the results? Can I possibly integrate these techniques into a product?
45-minute
Talk
Maximilian Baust - Senior R&D Engineer & Research Specialist, Konica Minolta Laboratory Europe
E8346 -
Object Detection Training: An Online Learning Pipeline for Humanoid Robots
This talk will feature the visual system we devised to train, within a few seconds, our R1 humanoid robot to detect multiple novel objects in a given scene. We will present our recently proposed on-line detection pipeline, which combines Faster R-CNN with one of the fastest kernel based methods, FALKON, to efficiently address the computationally challenging task of training an object detector in few seconds. By relying on a quick bootstrapping approach, the proposed algorithmic solution provides a 60x speedup in training with respect to standard region based methods, and is also in terms of prediction accuracy. Quantitative results will be shown with a demonstration of the system, which can exploit GPU acceleration either from a Geforce GTX 1080 Ti or two Jetson TX2 cards on board R1.
45-minute
Talk
Elisa Maiettini - Ph.D. Student, Istituto Italiano di Tecnologia Giulia Pasquale - Postdoctoral Researcher, Istituto Italiano di Tecnologia
E8356 -
Infrastructure choices for machine learning and deep learning- Presented by CISCO
In this session, you will learn what factors to consider when making infrastructure choices and what the benefits of on-premise infrastructure are for machine learning and deep learning. When considering artificial intelligence and machine learning, model development and algorithm choices are key. But anybody who attempted to train a model on a full dataset will tell you that infrastructure matters, a lot in fact. You can save hours, days, and sometime weeks by running your training algorithm on the right GPU-accelerated infrastructure. On the other hand when hunting for the perfect model for your business problems, you need to be able to stand up infrastructure quickly, connect to the existing data lake and iterate through many, many algorithm variations.
45-minute
Talk
Vikas Ratna - Product Management, Cisco Ravi Mishra - Sr. Technical Marketing Engineer, Cisco Systems Inc
E8437 -
Next Generation VR with Large FoV and High Resolution with StarVR and NVIDIA VRWorks
We will talk about the implementation of new features in VR and illustrate examples of how StarVR combined with NVIDIA technology is finally providing a natural life like immersion revolutionizing VR applications for industries as various as design, manufacturing, training or film and gaming. We will also cover early integrations in UE4,VRED,Techviz. The main drawbacks of current generation VR headsets are the narrow field of view (FoV) and the low perceived resolution resulting on a visible pixel grid which, combined, alter the immersive feeling for the user. StarVR is developing a new VR headset combining a large FoV of 210° on the Horizontal axis and 130° on the Vertical axis and two high resolution OLED panels with a combined resolution of 5K, aiming for an improved VR experience. However, rendering high quality VR content with a high and constant framerate to provide a smooth experience is a great technological challenge for graphics cards. To tackle that problem, StarVR has combined VRWorks technologies such as Quadro VR SLI for rendering one eye per GPU and MRS/LMS with the embedded Tobii Eyetracker resulting in a huge boost of performance with Pascal and Volta based GPUs
20 Minutes
Talk
Marc Piuzzi - VR Solution Engineer, STARVR CORP
E8486 -
ADAS in Sports Cars
Porsche's view on ADAS, with a focus on the specific profile that Porsche sees for sports cars. Porsche's experience with advanced systems like Innodrive as well as topics like the demands on V&V and data driven development.
45-minute
Talk
Juergen Bortolazzi - Dr. Ing. h.c. F. Head of ADAS, Porsche AG
E8513 -
Learn How the New NVIDIA vGPU Software Helps Deliver the Agile Data Center
With the latest release of NVIDIA vGPU software the world's most powerful virtual workstation gets even more powerful. Learn more about how our latest enhancements enable your data center to be more agile and scale your data center to meet the needs of thousands to ten-thousands and even hundreds of thousands of users. The newest release of NVIDIA virtual GPU software adds support for more powerful VMs, which can be managed from the cloud or from the on premises data center, or private cloud. With support for live migration of GPU-enabled VMs, IT can truly deliver high availability and a quality user experience. IT can further ensure they get the most out of their investments with the ability to re-purpose the same infrastructure that runs VDI during the day to run HPC and other compute workloads at night. In this session, we will unveil the new features of NVIDIA vGPU solutions and demonstrate how GPU virtualization enables you to easily support the most demanding users and scale virtualized, digital workspaces on an agile and flexible infrastructure, from the cloud and as well as the on premises data center.
45-minute
Talk
John Fanelli - VP of Product Management, NVIDIA Anne Hecht - Senior Director of Product Marketing, NVIDIA
E8517 -
Investing in XR. Where's the smart money going?
Vive X is HTC's global program to build the XR ecosystem by investing in startups. With over 80 portfolio companies and local hubs in the US and Asia, the program is now active in Europe, where over 150 startups have been looked at in the last six months. This talk will share the insights gained from this activity as well as discussing the latest investment trends in the sector generally. Who is investing in AR and VR and what types of ventures are most successful in getting funded? Where do the biggest opportunities lie and what are some key challenges that still need to be addressed? This talk is for you whether you're a startup founder, investor or just interested in the latest innovations happening in XR.
20 Minutes
Talk
Dave Haynes - Director, Vive X Europe, HTC
The NVIDIA Docker plugin makes it possible to containerize production-grade deep learning workflows using GPUs. Learn to reduce host configuration and administration by:
· Learning to work with Docker images and manage the container lifestyle
· Accessing images on the public Docker image registry—DockerHub—for maximum reuse in creating composable lightweight containers
· Training neural networks using both TensorFlow and MXNet frameworks
Upon completion, you’ll be able to containerize and distribute pre-configured images for deep learning.
Prerequisites: Basic experience with a bash terminal
When a trained neural network is tasked to find the answer on new data inputs, it is referred to as deployment. TensorRT is the primary tool for deployment, with various options to improve inference performance of neural networks. In this mini course, you'll:
· Learn how to use giexec to run inferencing
· Use mixed precision INT8 to optimize inferencing
· Leverage custom layers API for plugins
Upon completion, you'll know how to use TensorRT to accelerate inferencing performance for neural networks.
Prerequisites: Basic experience with CNNs and C++
90 minutes
DLI Instructor-Led Training
Adam Thompson - Sr. Solution Architect, NVIDIA
Get a hands-on practical introduction to deep learning for radiology and medical imaging. You'll learn how to:
· Collect, format, and standardize medical image data
· Architect and train a convolutional neural network (CNN) on a dataset
· Use the trained model to classify new medical images
Upon completion, you’ll be able to apply CNNs to classify images in a medical imaging dataset.
Prerequisites: None
90 minutes
DLI Instructor-Led Training
Nicola Rieke - Senior Solutions Architect, NVIDIA
E8198 -
Deep Learning Toolkit for Medical Imaging (DLTK)
DLTK is an open-source toolkit providing baseline implementations for efficient experimentation with deep learning methods on biomedical images. DLTK builds on top of TensorFlow, and its high modularity and easy-to-use examples allow for a low-threshold access to state-of-the-art implementations for typical medical imaging problems. Automatic downloading and pre-processing of example datasets allow for running and testing example applications, including medical image segmentation, regression, classification, representation learning, super-resolution and training generative models on biomedical images.
A comparison of DLTK's reference implementations of popular network architectures for image segmentation demonstrates new top performance on the publicly available challenge data "Multi-Atlas Labeling Beyond the Cranial Vault". Additionally, DLTK contains a medical model zoo with downloadable pre-trained models for medical image analysis problems, enabling transfer-learning and direct deployment of evaluated deep learning methods.
45-minute
Talk
Martin Rajchl - IC Research Fellow, Imperial College London
E8251 -
Predicting Atmospheric Turbulence: The Key to Imaging Habitable Planets with Large Telescopes
Come and learn how GPUs help identify biological activity on nearby exoplanets. Deployed on on the Japanese Subaru telescope at 4,200m elevation atop Maunakea, Hawaii, the GPU hardware technology constitutes the backbone of the adaptive optics, which drives the real-time correction of the optical aberrations introduced by Earth's atmosphere. Using machine learning technique and advanced linear algebra algorithms accelerated by GPUs, a predictive control problem can now be solved at the multi-kHz frame rate required to keep up with turbulence changes. This represents the first successful on-sky result of this approach for exoplanet imaging.
45-minute
Talk
Olivier Guyon - Associate Professor, University of Arizona & Subaru Telescope Damien Gratadour - Associate Professor, LESIA, Observatoire de Paris
E8366 -
What is Needed to Industrialise Deep Learning?
At RoboVision we built an industrial pipeline that starts from scalable data curation to on premise deployment. Our goal is to put the internal teams at the customer in the driving seat, making them able to generate powerful deep learning models, deploying them, and getting more value from their DGX1 investments, without lengthy consultancy. With android integration and predictive labeling we enable a big crowd to annotate data, directly accessible for multi-gpu deep learning sessions. The data is stored on high speed storage systems like Pure. The tool uses nifty techniques like predictive labeling and multi-user curation to guarantee high quality input for the configurable deep learning stack. This stack is made scalable and robust, with the help of Kubernetes and MySQL clusters, ending in a restful API system for rapid integration in the client's ecosystem. After pioneering diverse applications in agriculture, RoboVision expanded its focus and is now active in industrial automation, safety and security, smart city and surveillance markets.
45-minute
Talk
Jonathan Berte - CEO, RoboVision
E8400 -
Collaborative Architectural Design Across Continents at KPF with NVIDIA Holodeck
During this session, we will present our development and exploration of multi user collaborative VR, using NVIDIA Holodeck, on large international projects. KPF has 7 global offices working as one large firm and with the increase for high level collaboration, we are continuously looking for ways to improve communication while reducing meeting and travel times. Testing remote collaborative VR environments to ensure our designs are communicated efficiently between teams and our stakeholders are especially valuable during the early fast iterative design stages of a project. This presentation should give you a real world overview of test cases for NVIDIA Holodeck on global projects.
45-minute
Talk
Cobus Bothma - Director - Applied Research, KPF
E8417 -
How to Create Talented Teams and Deliver Successful AI Solutions
This panel will showcase how management teams can implement new AI/DL solutions quickly and effectively, by developing talented teams successfully. It will also discuss how POCs can be moved seamlessly into productive use.
45-minute
Panel
Stich Timo - Senior Research Scientist, ZEISS Will Ramey - Sr. Director Global Head of Developer Programs, NVIDIA Micael HOLMSTROEM - Managing Director, Academy.se Ulli Waltinger - CT Research in Digitalization and Automation, Siemens
E8430 -
Large Batches for Faster Deep Neural Network Training on GPU
This talk demonstrates how Large Batch Training can be used to significantly speed up the training of deep neural network models on GPUs. The technique relies on recent results suggesting rules for scaling batch size and learning rate schedules. We apply Large Batch Training on deep neural networks that arise from problems in quantitative finance. Our experimental data shows that Large Batch Training scaling rules remain valid in the financial domain, supporting empirical evidence in the literature, and can be used to exploit massive parallelism offered by GPUs for deep neural network training.
45-minute
Talk
Jonathan Kochems - Quantitative Research, JP Morgan Baranidharan Mohan - Quantitative Research, JP Morgan
E8450 -
Towards Understanding Deep Neural Network Behaviour
Interpretability of deep neural networks has focused on the analysis of individual samples. This can distract our attention from patterns originating at the distribution of the dataset itself. We broaden the scope of interpretability analysis, from individual images to entire datasets, and found that some high-performing classifiers use less than half the information contained in any given sample. While the learned features are more intuitive to visualize for image-centric neural networks, in time-series it is much more complicated as there is no direct interpretation of the filters and inputs as compared to image modality. In this talk we are presenting two approaches to analyze the behavior of networks with respect to used input signal, which pave the way to go beyond simple layer stacking and towards a more principled design of neural networks.
45-minute
Talk
Sebastian Palacio - PhD Researcher, DFKI
E8461 -
How Robots Can Learn to Master Dynamic, Complex Tasks through Human Interaction
This talk is about driving industrial robots from Jetson TX2: The difficulties of getting pixels from two industrial cameras into the system and a control signal out at high frequencies and under real-time conditions. micropsi industries MIRAI is a machine learning-driven real-time control system for industrial robots developed on x86, and the talk will relate the experiences made when getting this to run on Tegra: numbafying CPU-intensive algorithms, learning the quirks of Tegra CPU management, going from TensorFlow to a hand-rolled TensorRT integration and back — and doing all this in Python.
45-minute
Talk
Ronnie Vuine - CEO, micropsi industries
E8465 -
Multi-Sensual Experiences Based on VR Technology at Audi DesignCheck
The presentation gives a deep insight into the virtual development stages at Audi design check. It shows how we react to new customer wishes and how we virtually design, conceptualize and develop the interior of the future. In the past we have focused on 2d power walls. Today we have the possibility to move freely in interactive rooms and to make the car experienceable. To drive on virtual roads or to communicate in collaborative spaces by using 3d data. Starting from the concept phase to the start of production. An impressive leap forward. Vorsprung durch Technik.
45-minute
Talk
Daniel Hauser - Projectmanager Virtual Reality, Audi AG
E8469 -
AI in HD mapping.
By now the industry agrees that HD maps are needed for autonomous driving. Cars need to position themselves very accurately and be aware of the road ahead in order to plan their next move. In this panel session on HD mapping, three map companies will talk about how they are building HD maps in different regions of the world and how to automate map making using AI. But even more important, how will they keep their HD maps up to date? After all, an out of date HD map will not help the car. The panel will also touch on how cars should access the latest, most up-to-date HD maps with minimal latency.
45-minute
Panel
Willem Strijbosch - Head of Autonomous Driving, TomTom Ralf Herrtwich - Senior Vice President Services Group, HERE Technologies William Raveane - Lead AI Engineer, Navinfo Europe Bahram Yoosefizonooz - Technical Director, Navinfo Europe
Project Hydrogen: Unifying State-of-the-Art AI and Big Data in Apache Spark
Data is the key ingredient to building high-quality, production AI applications. It comes in during the training phase, where more and higher-quality training data enables better models, as well as during the production phase, where understanding the model’s behavior in production and detecting changes in the predictions and input data are critical to maintaining a production application. However, so far most data management and machine learning tools have been largely separate. In this presentation, we’ll talk about several efforts from Databricks, in Apache Spark, as well as other open source projects, to unify data and AI by using accelerated hardware such as GPUs. Through a demo, we will show how this project significantly simplifies AI applications at scale.
1 Hour
Workshop
Tim Hunter - Software Engineer, Databricks
E8177 -
How GPUs Speed the Analysis of Real-Time Risk, Fraud Detection, and Trader Surveillance
Find out how financial companies analyse data using NVIDIA GPU-acceleration, impacting real-time risk management, regulatory reporting, fraud detection and cybersecurity, anti-money laundering, and trader surveillance.
We will discuss real-world examples, including how a specific multinational bank uses a real-time risk management engine running on GPU cloud instances. The bank's analysts can now make time-sensitive, computation-intensive risk calculations involving hundreds of variables, using a real-time, interactive dashboard. This produces meaningful, timely, and consistent financial analysis, optimised to maximise profitability and power business in motion.
This approach lends itself to a data-powered business. It allows banks to move applications – such as counterparty risk analysis – from batch overnight processing to streaming and real-time, creating flexible real-time monitoring of extreme data that makes it easy for traders, auditors, and management to take action.
45-minute
Talk
James Mesney - Principal Solution Engineer, KInetica
AI makes it possible for robots to perceive and interact with their environments, enabling them to take on tasks that were unthinkable--until now. NVIDIA ISAAC accelerates development and deployment with an advanced SDK and tools for enhanced robotics simulation. Claire Delaunay VP of Engineering at NVIDIA will present on the ISAAC SDK and GEMS, while Dieter Fox will discuss Robotics research projects
2 Hour
Workshop
Dieter Fox - Senior Director of Robotics Research, NVIDIA Claire Delaunay - VP of Engineering at NVIDIA, NVIDIA
Learn how to accelerate your C/C++ or Fortran application using OpenACC to harness the massively parallel power of NVIDIA GPUs. OpenACC is a directive-based approach to computing where you provide compiler hints to accelerate your code, instead of writing the accelerator code yourself. Get started on the four-step process for accelerating applications using OpenACC:
· Characterize and profile your application
· Add compute directives
· Add directives to optimize data movement
· Optimize your application using kernel scheduling
Upon completion, you will be ready to use a profile-driven approach to rapidly accelerate your C/C++ applications using OpenACC directives.
Prerequisites: Basic experience with C/C++
90 minutes
DLI Instructor-Led Training
Issam Said - Solution Architect, NVIDIA
E8140 -
World Models and AIs That Invent Their Own Goals
Since 2009, our deep learning artificial neural networks have won numerous contests in pattern recognition and machine learning. Today, they are used billions of times per day by the world's most valuable public companies. True AI, however goes far beyond slavishly imitating teachers through deep learning. That's why we have also focused, since 1990, on unsupervised AIs that invent their own goals and experiments to figure out how the world works and what can be done in it. Many of them model the world through a recurrent neural network that learns to predict the consequences of their action sequences. Without a teacher, they derive rewards from continually creating and solving their own, new, previously unsolvable problems, a bit like playing kids do, to become more and more like general problem solvers in the process. Relevant buzzwords include "artificial curiosity" (since 1990) and PowerPlay (since 2011). I will also briefly outline how AIs that set their own goals will eventually colonise the entire universe and make it intelligent.
45-minute
Talk
Jürgen Schmidhuber - Scientific Director of IDSIA, IDSIA
Unlike other industries, architecture has still a lot of frontiers to adopt digital tools, impacting productivity by the use of limited and disjointed solutions. While being used by big architecture and automotive companies, Dassault Systèmes 3DEXPERIENCE has recently expanded its breadth to Cloud licensed users, giving a much wider access to the latest CATIA features including improved visualization on large datasets and native VR integration with NVIDIA GPU. This talk will present how SHoP leverages CATIA for engineering, wireframing and building systems automation but also their early investigations with latest R2018x built-in VR immersive experiences for design validation and collaborative decision workflow.
45-minute
Talk
Stephan Ritz - CATIA Design, Product Experience | Portfolio Management Director, Dassault Systemes Geoffrey Bell - Associate, Digital Design and Construction, SHoP Architects
Remove From ScheduleAdd To ScheduleAre you sure you would like to Delete this personal time?Edit My ScheduleEdit Personal TimeThis session is full. Would you like to be added to the waiting list?Would you like to remove "{0}" from your schedule?Would you like to add "{0}" from your schedule?Sorry, this session is full.Waitlist AvailableSorry, this session and it's waiting list are completely full.Sessions AvailableAdding this multi-day session automatically enrolls you for all times shown below.Removing this multi-day session automatically removes you for all times shown below.Adding this multi-day session automatically enrolls you for all session times for this session.Removing this multi-day session automatically removes you for all session times for this session.Click to view detailsInterestsHide InterestsSearch SessionsExport ScheduleThere is a scheduling conflict. You cannot add this session to your schedule because you are participating in another session at this time.Schedule Conflict.An error occurred while processing this request..
Adding this item creates a conflict with another session on your schedule.Remove from Waiting ListAdd to waiting listRemoving this will remove you from the waiting list for all session times for this sessionAdding this will add you to the waiting list for all session times for this session.You have nothing scheduledTap below to see a list of sessions and activities that are available to add to your schedule this weekChoose from the list of sessions to the left to add to your schedule for the dayAdd a Session
We use cookies on this website to enhance your browsing experience and measure our audience. Click here to find out more about how we use cookies. By continuing to use this website, or by closing this box, you are indicating your consent to our use of cookies.