This site requires Javascript in order to view all its content. Please enable Javascript in order to access all the functionality of this web site. Here are the instructions how to enable JavaScript in your web browser.

NVIDIA at KDD 2020

Data science (DS), machine learning (ML), and deep learning (DL) yield exciting new capabilities for researchers. But, using multiple tools that don’t always work together seamlessly can add layers of complexity. With RAPIDS, a software accelerator from NVIDIA that speeds up DS workflows with GPUs, scientists can use the same tools for DS, ML, and DL—and deliver results faster.

HIGH PERFORMANCE DATA SCIENCE FOR THE ENTERPRISE

NVIDIA is committed to simplifying, unifying, and accelerating open source DS. By optimizing the entire stack—from hardware to software—and by removing bottlenecks, we enable data scientists to do more than ever, with much less.

Using RAPIDS, scientists can focus on seamless integration with familiar APIs and interoperability with the broader open-source DS community. They can build complete end-to-end pipelines on RAPIDS, including ETL, feature engineering, ML/DL modeling, inference, and visualization—all while removing typical serialization costs and providing speedups of up to 50X.

Sessions

Keynote: Next-generation frameworks for Large-scale AI

Hear Anima Anandkumar, director of Machine Learning Research at NVIDIA, discuss the availability of web-scale labeled data and parallelism of GPUs that enable users to harness the power of neural networks. Further progress requires reducing dependence on labeled data, and design algorithms that can incorporate more structure and domain knowledge. Learn about efficient frameworks that enable developers to easily prototype such models to incorporate tensorized architectures.

Attend Keynote

Tutorial: Accelerating and Expanding End-to-End Data Science Workflows with DL/ML Interoperability Using RAPIDS

Join NVIDIA engineers and data scientists as they walk through hands-on DS, ML, and DL engineering problems and use cases. They’ll demonstrate how RAPIDS running on Microsoft Azure ML can be used for end-to-end, entirely-GPU pipelines. The tutorial also includes specifics on using RAPIDS for feature engineering, interoperability with common ML/DL packages, and creating GPU-native visualizations using RAPIDS cuXfilter. The tutorial content is open-sourced and can be found here.

See Tutorial

Workshop: NVIDIA Merlin - A GPU-Accelerated Recommendation Framework

Hear Even Oldridge from NVIDIA’s RecSys engineering team present on NVIDIA Merlin, a framework for building high-performance, DL-based recommender systems, during the first International Workshop on Industrial Recommendation Systems on August 24, 2020 from 10:40-11:00 a.m. PDT.

Data Science Solutions

Increase Model Accuracy with Hyperparameter Optimization

ML models can have dozens of options, or hyperparameters, that can greatly affect model accuracy. Accelerated ML models in RAPIDS allow you to use hyperparameter optimization (HPO) experiments to identify the most accurate model for your problem. RAPIDS supports HPO solutions based on AWS Sagemaker, Azure ML, Google Cloud AI, Dask ML, Optuna, and Ray Tune frameworks, so you can easily integrate with whichever framework you use today.

Learn More

Boost Revenue with Recommender Systems

Recommendation systems drive every action that consumers take online. On some of the largest commercial platforms, recommendations account for as much as 30% of the revenue. A 1% improvement in the quality of recommendations can translate into billions of dollars. Use NVIDIA Merlin, a framework for building high-performance, DL-based recommender systems, to speed up the entire pipeline, from ingesting and training to deploying GPU-accelerated recommender systems.

Simplify Conversational AI Deployment

NVIDIA now makes it easier to deploy conversational AI. Neural Modules (NeMo), a new open source toolkit, enables developers to simply compose complex neural network architectures for conversational AI. Once the model is built and trained, NVIDIA TensorRT can optimize performance and ease deployment. The NVIDIA Jarvis SDK also allows developers to build and deploy AI applications that fuse vision, speech, and other sensors.

Learn More

GTC 2020

Innovation never sleeps. Experience the latest breakthroughs online this fall.

WATCH VIDEO

LIKE NO PLACE
YOU’VE EVER WORKED

You’ll solve some of the world’s hardest problems and discover never-before-seen ways to improve the quality of life for people everywhere. From healthcare to robots. Self-driving cars to blockbuster movies. And a growing list of new opportunities every single day. Explore all of our open roles including internships and new college graduate positions.

FIND YOUR NEXT JOB