Artificial intelligence is arguably the most important foundational technology being developed in the world today. The field of AI has become enormously popular with students at colleges and universities, and companies everywhere are opening AI departments to capitalize on the massive potential of this technology. Looking to the future, we need an AI roadmap to meet the needs of the rapidly growing number of individuals and organizations seeking to build and access AI solutions.
This is particularly true when it comes to meeting the needs of enterprise AI, which has important differences from consumer AI. For example, consumer-oriented AI has become very good at training on extremely large sets of labeled data to perform very narrow tasks (e.g., locating a specific category of images in pictures). But for AI to truly benefit the enterprise, it must be able to be trained on smaller sets of labeled data and learn to handle whatever task or business process is at hand. In many cases, users of enterprise AI will also need to explain why an AI model produced a certain outcome. For example, doctors and clinicians using AI systems to support medical decision making must be able to provide specific explanations for a diagnosis or course of treatment, both for regulatory and liability reasons. We must have a better theoretical and grounded understanding of what is happening inside these machine learning systems. Research is ongoing to make AI less of a black box and develop the appropriate algorithms and models that make enterprises of all kinds successful.
Dario Gil explores the challenges faced by companies building AI solutions for enterprise applications and areas of research required to drive this field forward. IBM Research AI is developing techniques to work with existing AI models to build custom, high-accuracy, domain-specific models faster, using less data. These techniques speed the training process when entering a new domain and enable AI systems to learn on smaller data. Through innovations in the physics of AI, IBM Research teams are investigating new materials, devices, and architectures for analog AI computation, as well as the intersection of quantum computing and machine learning, that will eventually realize orders-of-magnitude improvements for AI computational efficiency.
This session is sponsored by IBM Watson.
Dario Gil is a leading technologist and senior executive at IBM. As vice president of AI and IBM Q, Dario is responsible for IBM’s artificial intelligence research efforts and for IBM’s commercial quantum computing program, IBM Q. Previously, he was the vice president of science and solutions, directing a global organization of 1,500 researchers across 12 laboratories with a broad portfolio of activities spanning the physical sciences, the mathematical sciences, and industry solutions based on AI, IoT, blockchain, and quantum technologies. His research results have appeared in over 20 international journals and conferences, and he is the author of numerous patents. Dario is an elected member of the IBM Academy of Technology. He holds a PhD in electrical engineering and computer science from MIT.
Comments on this page are now closed.
©2018, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • email@example.com