UNH SAIL Lab (@UNHSAIL) header
UNH SAIL Lab (@UNHSAIL) profile picture

The Secure and Assured Intelligent Learning (SAIL) lab - University of New Haven PI: @vbehzadan

Joined June 2020
Photo by UNH SAIL Lab (@UNHSAIL)
New #preprint: Inspired by adversarial examples, we explore the vulnerability of EEG-based BCIs to Adversarial Stimuli -minor perturbations applied directly at the sensory level. In experiments w. motor imagery tasks, all participants were affected by minor visual perturbations.
Replying to @arxiv_cs_LG
Adversarial Stimuli: Attacking Brain-Computer Interfaces via Perturbed Sensory Events. Bibek Upadhayay and Vahid Behzadan arxiv.org/abs/2211.10033
2
4
Replying to @mh_rahmani
@mh_rahmani demonstrating his wizardry powers at @UNHSAIL lab. #bci #Neurotech
Adversarial ML meets Wall Street: In this week’s SAIL seminar, My PhD student Nancirose Piazza presented our research on adversarial manipulation of RL-based algorithmic trading. youtu.be/aF6yX9AfnnI
1
1
Many thanks to the RiskLab Middle East (@ Sharif University) for inviting me to present @UNHSAIL 's recent research on the security of deep algorithmic trading policies today. The recording of my talk will (hopefully) be posted soon.
1
4
Received that unhappy "we do regret" line from #aaai2021? Do not despair! If your paper fits within the #privacy in #AI/#ML and/or #policy scope, submit it to the PPAI workshop (AAAI Fast Track)! Deadline: December 7, 2020, AOE Check out ppai21.github.io for more info!
6
11
UNH SAIL Lab retweeted
Congratulations to @MilanZinzuvadiy for successfully defending his master's thesis today. All of us at @UNHSAIL are very proud of Milan's achievements, not only because his thesis work will be presented at #aaai2021, but also because Milan is the first graduate SAILor to defend.
SAIL has not 1, but 2 talks @ #defcon28: @levitannin will talk about "Ambly, the Smart Darknet Spider" at @ReconVillage (8/8-1300) & Dr. @vbehzadan will present "Faults in our Pi Stars: Security Issues in Deep Reinforcement Learning" at @aivillage_dc (8/9-1300). #SAILStrong
Check out this interview the Director of SAIL Lab, @vbehzadan with ABC News about his work on a Psychopathological Approach to AI Safety #aisafety soundcloud.com/vahid-behzada…
1
1