By happy circumstance, Santa Clara-based chip maker NVIDIA finds itself in the position of being an artificial intelligence (AI) startup king maker. The company designs and manufactures the entire computing platform for deep learning, the fastest growing field in AI, building everything from graphics processing units (GPUs) to software to systems purpose-built for deep learning. As the name might suggest, GPUs were developed to improve the computer graphics experience by offloading certain computationally intense image processing tasks from the standard central processing unit (CPU). The particular strength of a GPU is performing large numbers of parallel floating point calculations. This helps computer screens to increase in detail and complexity without sacrificing system performance, and modern gaming would not be possible without it.
Then, in 2012, Andrew Ng and a group of researchers at Stanford University discovered that GPU chips designed for video games could greatly speed up the training of neural network nodes. Ng found that a cluster of GPUs could accomplish the same task in a day that would take weeks for traditional CPU-based methods of neural net processing. Benchmarks have since found that a single DGX-1 (NVIDIA’s latest deep learning supercomputer) node can outperform 250 CPU-based servers.
Since then, NVIDIA has actively promoted deep learning technology. Its latest effort is the NVIDIA Inception Program. The program provides tools, technical resources, and marketing opportunities for startups. Among the benefits NVIDIA offers entrepreneurs:
- Deep learning problem solving expertise – Access to NVIDIA’s deep learning experts and dedicated engineering teams.
- NVIDIA’s global partner network – Exposure to NVIDIA customers, partners, and suppliers.
- Technical training – Online and in-person courses via the NVIDIA Deep Learning Institute.
- Funding – Qualified companies may be eligible for seed funding through NVIDIA’s GPU Ventures Program.
“For companies seeking funding, being involved in a program like Inception is a plus,” says Neville Teagarden, CEO at AI Capital, a venture fund devoted to startups using AI technology. “The ability to get technical support alone shows the founders are serious about bringing a solution to market faster.”
Yet the greatest benefit for the new companies may not be in raising funds but in finding hardware, marketing, partners, and finally an exit. Every spring, the company holds its GPU Technology Conference. As part of the event, NVIDIA conducts an Emerging Companies Summit, which showcases promising companies that use GPUs. Past participants included Oculus Rift (acquired by Facebook for $2 billion), Gaikai (acquired by Sony for $380 million), and Natural Motion (acquired by Zynga for $527 million). This year, participant Magic Pony was acquired by Twitter for an undisclosed sum.
“The majority of AI startups need access to technology and experts. This is where Inception adds the most value, acting as an incubator during a critical stages of product development, building demos or scaling to deployment”, says Kimberly Powell, Senior Director of Industry Business Development at NVIDIA and the head of the program. “Helping startups reach their milestones and introducing them to opportunities like Emerging Companies Summit is what Inception is all about.”
The initial application form is simple, asking only nine general questions and one qualifying question about how GPUs will help build a company’s products and services. According to Powell, more than 600 companies have already applied and the company is actively contacting them to build a more robust ecosystem around its technology.