Nvidia Corporation: At the Forefront of AI Revolution

Follow Nvidia Corporation on :

Jensen Huang, Founder & CEO

The burgeoning ecommerce industry is putting enormous pressure on the world’s transport systems. Just consider these numbers: more than 35 million packages across the world are being delivered each day, a number that is mushrooming by 28 percent annually. By the year 2040, delivery services will need to travel an additional 78 billion miles per year to handle goods ordered online, according to research conducted by KPMG.

With consumers expecting overnight or same-day deliveries, it’s far to state that transportation companies have never been under more duress. Luckily for them, autonomous trucks are arriving just in time to meet the rapidly-escalating demands. Since autonomous trucks can operate 24 hours a day, they are bound to improve delivery times and increase the efficiency of deliveries. Autonomous trucks are capable of automating short, routine trips like the loading and/ or unloading of containers on cargo ships, managing port operations, and autonomously driving on the freeway. Experts believe they can dramatically streamline the shipping industry. According to a McKinsey report, autonomous trucks have the potential of reducing the annual cost of logistics in the US by 45 percent—between $85Bn and $125Bn.

Brought together by a common goal to deliver “autonomy” to the world’s transportation industries, NVIDIA and Volvo Group have recently entered a strategic partnership. NVIDIA—the global leader in applying AI and deep learning to high-performance computing, healthcare, and self-driving cars—and Volvo Group, one of the leading truck manufacturers in the world, plan to use AI to revolutionize how people and products move all over the world.

As part of the partnership, Volvo Group will leverage the power of the NVIDIA DRIVE end-to-end autonomous driving platform to test, train, and deploy self-driving AI vehicles at its headquarters in Gothenburg, Sweden. The benefits of self-driving AI vehicles extend to industries such as freight transport, public transport, construction, mining, forestry, and collection of recyclable waste. By injecting AI into these industries, Volvo Group and NVIDIA believe they can create revolutionary vehicles that can deliver ultra-productive services.

Jensen Huang, the founder and CEO of NVIDIA, believes the alliance is another step toward AI driving the future of transportation. “Our two industries have been separate since their founding. Today we announced one partnership to develop the future together. For the first time, we can imagine supplying AI for something that wasn’t possible before, the automation of transportation,” says Huang.

By co-locating engineering teams in Silicon Valley and Gothenburg, NVIDIA and Volvo Group will build on the DRIVE AGX Pegasus platform for in-vehicle AI computing and use the full DRIVE AV software stack for 360-degree sensor processing, map localization, perception, and path planning. The companies will also test and validate the systems using the NVIDIA DRIVE hardware-in-the-loop simulation platform. “If we are to succeed in the future with speed, quality, and safety, and gain benefits of autonomous driving, we need to partner up with the best,” says Martin Lundstedt, the CEO of Volvo Group, on the partnership with NVIDIA. To ensure safe and reliable self-driving AI vehicles, Volvo Group will also be utilizing the NVIDIA DRIVE Constellation—a VR autonomous vehicle simulator—to test and validate AVs, so the trucks can handle various operating challenges across the world.

The application of AI in self-driving vehicles is just the tip of the iceberg, as far as NVIDIA is concerned.

NVIDIA’s GPU-accelerated systems are assisting government authorities during natural disasters, serving communities by improving safety, and enhancing power efficiency and traffic flows

In healthcare, the AI solution provider uses GPU computing to enhance a number of processes, ranging from medical imaging, analyzing genomes, to discovering new drugs. While financial companies leverage NVIDIA’s GPU-accelerated deep learning to improve fraud detection, large enterprises are enhancing their cybersecurity courtesy NVIDIA’s AI solutions. Meanwhile, digital customer services are benefitting significantly from the use of deep learning for image recognition to curate richer customer experiences for their users.

NVIDIA impacts a number of industries through patented products such as NVIDIA AGX SYSTEMS for self-driving cars, intelligent machines, and medical imaging, Tesla Data Center for GPUs, and the NVIDIA HGX-2, the world’s most powerful accelerated server platform for deep learning, machine learning, and HPC. NVIDIA also offers NGC, a GPU-optimized software that allows data scientists to build solutions, gather insights, and deliver business value.

AI for the Greater Good

However, NVIDIA’s AI solutions aren’t restricted to the business world. The Silicon Valley-based company is teaming up with educators, scientists, and humanitarians, to democratize AI and deep learning for the common good of humanity. NVIDIA’s GPU-accelerated systems are assisting government authorities during natural disasters, serving communities by improving safety, and enhancing power efficiency and traffic flows. NVIDIA also provides crucial aid during natural crises by helping healthcare companies to develop customized medicines.

At the government level, NVIDIA works closely with the US Federal Government to enable its platforms with AI and deep learning for benefiting in various key areas. For example, last year, the US Government spent $306Bn responding to hurricanes, storms, fires, earthquakes, and other natural disasters. During the process, it utilized NVIDIA’s GPU-accelerated AI systems to keep first responders out of harm’s way and saving lives. The US government is also recognizing the value of smart cities powered by NVIDIA’s AI solutions. There is already ample proof that AI can predictive infrastructure designed to help cities improve public safety, efficiencies of power usage, and dramatically reduce traffic congestion.

It’s clear that the US Government plans to increasingly implement AI as a core technology to solve today’s and tomorrow’s challenges. Since NVIDIA GPU computing and NVIDIA® CUDA® are widely recognized as the leading solutions to reduce the cost of AI adoption, while also offering high performance and compatibility, it won’t be long until nationwide adoption of smart cities, industrial robots, and autonomous public transport. To quicken the development of AI-empowered solutions at the city level, the US Government has taken the initiative to support fundamental AI research at universities.

The United States Geological Survey (USGS) also stands to benefit in the long run. NVIDIA is already applying AI to resolve uncertainties in climate projects.

AI to Spot Clouds on the Horizon

Indeed, NVIDIA has managed to change the way climate researchers make their projections. Until recently, climate researchers relied upon decades-old software to conduct their analyses. Since the legacy software architecture is complex to update with latest methodologies, a group of researchers is starting from the ground up, writing a brand new climate model that utilizes AI, NVIDIA GPUs, and other software tools. The group, titled the Climate Modeling Alliance (CliMA), comprises of scientists from MIT, Caltech, NASA’s Jet Propulsion Laboratory, and the Naval Postgraduate School.

By building a new climate model from scratch, climate researchers can better account for small-scale environmental features such as cloud cover, rainfall, sea ice, and ocean turbulence. Since these variables are too geographically minuscule to be captured in climate models, researchers are using AI and incorporating the AI projections into the new climate model, to effectively reduce the uncertainties by half when compared to existing models.

To accelerate their climate model calculations, CliMA is using NVIDIA V100 Tensor Core GPUs both onsite and on Google Cloud. The researchers also plan to use Summit, the GPU-powered supercomputer that brings together several key NVIDIA technologies such as the Volta GPU and the high-speed NVLink interconnect.

Besides enhancing climate research methodologies, NVIDIA is also using AI and GPUs to accelerate scientific research for higher education. Through the NVIDIA Deep Learning Institute (DLI), the company offers hands-on training in AI to solve complex, real-world problems. A cloud-based training program, the DLI provides developers, data scientists, researchers, and students with self-paced online training. This includes workshops, downloadable course materials, and instructor-led team sessions. Several universities across the world have enrolled into the DLI program, which offers a certificate of competency to support professional growth.

Transforming Retail

NVIDIA is doing wonders by imparting knowledge on AI and deep learning at the academic/research level, but that hasn’t stopped the company from working cohesively with the world’s biggest corporations. Case in point: Walmart, the bigger retailer on planet Earth. The retail giant has always invested heavily in the technology quotient of its operations and became an early adaptor of AI through NVIDIA GPUs and RAPIDS data science software. By utilizing NVIDIA RAPIDS, the open-source suite of GPU-accelerated libraries, Walmart is able to tear through its massive-scale data analytics and machine learning. By pairing RAPIDS with few other technologies, Walmart is able to crunch data 100 times faster than traditional methods. Through high-end data crunching, Walmart is improving its product forecasting, supply chain management, and predictive consumer buying trends. After all, there is no better way for the retail giant to serve more than 265 million weekly customers.

According to Bill Groves, Walmart’s chief data officer, the retail company is using NVIDIA’s patented technologies to perform product forecasting of more than 500 million store-item combinations each week. “More compute power allows us to bring in more data and get better faster. So when the customer comes in the product they want is sitting on the shelf. That’s what we’re always working towards,” Groves says.

Walmart is planning to utilize NVIDIA technology to other use-cases such as logistics, supply chain, merchandising, and in-store eCommerce functions. By working together, NVIDIA and Walmart are integrating cutting-edge GPU technologies at a breakneck speed. This gives Walmart the advantage of being “a first mover” in the space since NVIDIA is yet to forge any other strategic partnerships with retail companies.

For NVIDIA, collaborating with Walmart means it can test its tools and technology against real-time production data from the world’s largest retailer. It’s clearly a win-win situation. Groves added, “Math and science hasn’t changed in 30 to 50 years. What has changed is the technology that enables it, so that we have opportunities that we didn’t have until recently.”

In the coming years, every segment of the modern economy will feel the impact of AI. From energy, transportation, and healthcare, to manufacturing, entertainment, education, IT, and government, the AI boom is imminent. NVIDIA is at the forefront of this AI revolution.

- Joe Phillip
July 26, 2019