Seminar Videos
Vive Center 2023 AR/VR Seminar (3/17) – Unique Identification of 50,000+ Virtual Reality Users from Head & Hand Motion Data
Speaker: Vivek Nair
Abstract: With the recent explosive growth of interest and investment in virtual reality (VR) and the so-called “metaverse,” public attention has rightly shifted toward the unique security and privacy threats that these platforms may pose. While it has long been known that people reveal information about themselves via their motion, the extent to which this makes an individual globally identifiable within virtual reality has not yet been widely understood. In this study, we show that a large number of real VR users (N=55,541) can be uniquely and reliably identified across multiple sessions using just their head and hand motion relative to virtual objects. After training a classification model on 5 minutes of data per person, a user can be uniquely identified amongst the entire pool of 50,000+ with 94.33% accuracy from 100 seconds of motion, and with 73.20% accuracy from just 10 seconds of motion. This work is the first to truly demonstrate the extent to which biomechanics may serve as a unique identifier in VR, on par with widely used biometrics such as facial or fingerprint recognition.
Bio:
ROAR Spring 2022 Seminar (4/22) – Semi Autonomous Vehicles
Abstract: Recently, an extensive amount of time, money, and resources have been allocated in finding fully autonomous solutions to existing problems. The most public showcase of autonomous technology appears in vehicles today. Similar to the Space Race, many corporations today are racing to have the first fully autonomous vehicle on the market. However, an alternative solution could be reached using semi-autonomous modes rather than developing a fully autonomous vehicles. The goal of this research is to demonstrate that semi-autonomous modes for vehicles can outperform solely human controlled vehicles and autonomous vehicles. Proving this can focus future development of autonomous vehicles around drivers. Rather than being disconnected from the user, the autonomous system will be an integral part of the design for future technology. The semi-autonomous agent built in Berkeley’s ROAR project was able to achieve better performance compared to both human-controlled agents and autonomous agents in multiple race scenarios.
ROAR Spring 2022 Seminar (2/18) – Reinforcement Learning
In this week’s ROAR Seminar, the Berkeley Reinforcement Learning team will present the latest results in training an end-to-end autonomous driving agent using neural networks to directly control a vehicle’s steering and breaking. The talk will start by giving an introduction to basic reinforcement learning setting to play Atari games, and will also cover techniques to make the training of autonomous driving agent more efficient for the robot control applications.
ROAR Spring 2022 Seminar (1/28) – The Indy Autonomous Challenge – a global autonomous competition with a $1.5 million dollar prize. — How University of Hawai’i, UCSD, and UC Berkeley students teamed up to go toe to toe with 9 international teams.
Abstract:
What happens when your code controls a race car at 100mph? The AI Racing Tech team has been pushing these boundaries…
Imagine The Dallara AV-21 rounds the lap breaking 100mph, you can hear the direct mount engine roar past in a precision turn as your high level controls algorithms execute perfectly. The team’s lidar processing algorithms work with the GPS data bounding the track edges as a keep out zone. You stand with the pitt-crew on the edge of the track holding your breath watching across multiple monitors with the rest of the team as you approach the passing zone to overtake the competition. At this point it’s all strategy, your team’s software is everything…
The first history making IAC was held on Oct. 23, 2021 at the Indianapolis Motor Speedway where 21 universities from 9 countries forming 9 teams competed for the $1 million grand prize, in October of 2021.
Indy Autonomous Challenge global university teams returned to action at CES in January of 2022 when they competed at the Las Vegas Motor Speedway in the first high-speed, head-to-head autonomous racecar competition IAC showcase.
The IAC taps into the power of incentive prize competitions to inspire the best and brightest in universities worldwide to eliminate barriers to innovation, overcome complex challenges, and increase public awareness of the transformational impact that automation can have to improve vehicle safety and performance.
UC Berkeley members Chris Battista, Cake Deng, and C.K. Wolfe will talk about their race experience, the challenges, recruiting, and what it will take to prepare for October 2022 to race again.