How can AI Revolutionize Transportation?

How can AI Revolutionize Transportation? USC Computer Science and DiDi Symposium & Luncheon on Smart Transportation

By Caitlin Dawson / Images: Kayly Luong

On June 19, 2019, more than 35 guests from USC and DiDi arrived at USC's University Club to discuss the future of communication, transportation and artificial intelligence.

The event was part of the the USC Computer Science Industry Affiliates Program (IAP), which provides engagement opportunities with USC's talented students, world-class faculty, and innovative researchers in an array of path-setting disciplines.

Find out more about USC's IAP program

As the world's leading mobile transportation platform, every day, DiDi receives over 100TB new data, processes more than 40 billion routing requests, and acquires over 15 billion location points for its 550 million users.

DiDi's Los Angeles team is currently focused on natural language processing.

"Smart cars should speak and understand human language, and people should be able to order services and solve their problems by speaking. Also, people who speak different languages should be able to communicate with one another." — Kevin Knight, DiDi chief scientist for natural language processing, and USC computer science professor (on leave).

"If car is smart enough to drive you, it should also be smart enough to talk to you in your language." — Kevin Knight, DiDi chief scientist for natural language processing, and USC computer science professor (on leave).

"We want the computer to be able to communicate with people, not only in helpful but also an entertaining way." — Nanyun Peng, USC computer science research assistant professor, expert in AI and creative language generation.

"The greyhound stopped to get a hare cut."

— A pun generated by an AI tool designed by Peng, based on insights from humor theory and presented at the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics.

Amittai Axelrod, staff scientist at DiDi, discusses innovative ways to conduct natural language processing without a parallel corpus.

DiDi has real-world AI problems and opportunities and USC has talent and technologies. We hope this is just a starting point for deeper collaborations." — Jieping Ye, VP of DiDi, Head of AI Labs.

Bistra Dilkina, USC computer science assistant professor, explores the intersection of machine learning and discreet optimization.

"Some of the best research goes from practice to theory. By tackling real-world problems, industry and researchers can solve deeper issues together." — Kevin Knight, DiDi.

"Industry has real challenges and data, so by collaborating, researchers can really make an impact in the real world." — Nanyun Peng, USC.

"We're in a real position to make a difference in AI and transportation, from infrastructure and vehicles to ride-sharing platforms." — Jieping Ye, VP of DiDi, Head of AI Labs.

Appreciate

Credits:

Created with images by DariuszSankowski - "navigation car drive" • noodle kimm - "untitled image"