The IAIFI has partnered with Northeastern’s Theorynet program to coordinate visits to high school classes in the Boston area.
Sign up for a classroom visit.
This activity is coordinated by the IAIFI Public Engagement Committee.
Winter Family Day
IAIFI was pleased to participate in MIT’s Winter Family Day on Saturday, February 25, 2023. We presented three demonstrations of a robot dog that has been trained using reinforcement learning, which trains machines to take action without explicit instructions or corrections. IAIFI Fellow Ge Yang and MIT undergrads delighted adults and children alike with their robot dog running, spinning, and hopping!
Such a fun day sharing our robot “dog” & how it was trained using reinforcement learning at the @mitopenspace #winterfamilyday! Thank you for having us and for all the great questions from the audience! pic.twitter.com/4Rqk01yUND— IAIFI (@iaifi_news) February 25, 2023
Thank you to @EpisodeYang and his team of undergrad students (pictured here), Sophie, Evan, and Prince, for leading our @mitopenspace #winterfamilyday demos! @AIVOInfo pic.twitter.com/ytjPIZWWFP— IAIFI (@iaifi_news) February 25, 2023
View Winter Family Day schedule
Cambridge Science Festival
Did you see us at the Festival? Complete our survey to tell us what you thought!
IAIFI was excited to participate in the 2022 Cambridge Science Festival with both an Open Lab on October 6, 2022 and as part of the Carnival on October 9, 2022. Both events included a galaxy zoo and a robot demonstration. All events were free and open to the public.
IAIFI’s galaxy zoo demonstration usedlabels created by citizen scientists to automate data analysis using machine learning to better understand complex data from our galaxy.
- Thursday, October 6, 2022, 1:00–5:00 pm: Open Lab at Ray and Maria Stata Center, outside Room 134
- Sunday, October 9, 2022, 12:00–4:00 pm: Carnival, Kendall/MIT Open Space
The galaxy zoo demonstration was an excellent opportunity for adults and children alike to better understand the ways that machine learning and AI can be used in conjunction with the physical sciences to better understand and interact with our world (and beyond!). And you could learn how to get involved as a citizen scientist with this exciting project!
Lina Necib, Assistant Professor at MIT, walks us through what the galaxy zoo is and how it uses input from the general public (like you!) to help researchers better understand galaxies as the amount of data we have continues to increase exponentially.
IAIFI’s quadruped robot demonstration illustrated the use of reinforcement learning, which trains machines to take action without explicit instructions or corrections.
- Thursday, October 6, 2022, 1:00–5:00 pm: Open Lab at Hockfield Court, outside Ray and Maria Stata Center
- Sunday, October 9, 2022, 12:00–4:00 pm: Carnival Robot petting Zoo, Kendall/MIT Open Space
This work also included the robot learning to adapt to variable conditions, making it more robust when deployed in the real world, which is a strategy expected to be relevant for making real-time adjustments in the operation of physics experiments to maximize their reach.
The quadruped robot demonstration was an excellent opportunity for adults and children alike to better understand the ways that machine learning and AI can be used in conjunction with the physical sciences to better understand and interact with our world (and beyond!).
We demonstrated two different robots–one that is programmed to move in a variety of ways and one that is programmed to play soccer!
Come see the robot show off for you in person! Will be here until 5:00 as part of the #CambridgeScienceFestival! pic.twitter.com/1W4Tewj79t— IAIFI (@iaifi_news) October 6, 2022
Robot soccer anyone? #cambridgesciencefestival pic.twitter.com/Vjw8fOqaIe— IAIFI (@iaifi_news) October 6, 2022
Recently, MIT and IAIFI researchers made headlines with the success of this training to get a mini cheetah robot to break speed records.
Engage with IAIFI
IAIFI is galvanizing AI innovation and advancing physics knowledge. Learn more about IAIFI, public events, and research.
Sign up for our mailing list and follow us on Twitter!