Goto

Collaborating Authors

 San Diego County


'What is the mission?' With Iran, California military families fear another 'forever war'

Los Angeles Times

Things to Do in L.A. With Iran, California military families fear another'forever war' Shalena Critchlow, at the Oceanside Pier, holds a photo of her son Cpl. Saiveon Critchlow, who recently completed his service with the U.S. Marines. This is read by an automated voice. Please report any issues or inconsistencies here .


"Reckless, Suicidal Race": The Deadly Threat Posed by Artificial Intelligence

Der Spiegel International

The rapid rise of artificial intelligence is just as dangerous as the invention of nuclear weapons. AI pioneers are warning that the technology must be brought under control while it's still possible. In November 2017, a short, eight-minute horror film went viral on the internet. The dystopic Slaughterbots" depicts attack drones equipped with facial recognition software infiltrating the U.S. Congress and universities and selectively killing senators and students who have spoken out against corruption. The film never makes clear who is behind the attacks. Or perhaps a cutting-edge arms manufacturer producing new AI-powered weapons, led by a young, charismatic CEO who markets autonomous killer drones as the future of warfare. The article you are reading originally appeared in German in issue 10/2026 (February 27th, 2026) of DER SPIEGEL. Slaughterbots" was made as a warning.


The Implicit Bias of Adam and Muon on Smooth Homogeneous Neural Networks

Gronich, Eitan, Vardi, Gal

arXiv.org Machine Learning

We study the implicit bias of momentum-based optimizers on homogeneous models. We first extend existing results on the implicit bias of steepest descent in homogeneous models to normalized steepest descent with an optional learning rate schedule. We then show that for smooth homogeneous models, momentum steepest descent algorithms like Muon (spectral norm), MomentumGD ($\ell_2$ norm), and Signum ($\ell_\infty$ norm) are approximate steepest descent trajectories under a decaying learning rate schedule, proving that these algorithms too have a bias towards KKT points of the corresponding margin maximization problem. We extend the analysis to Adam (without the stability constant), which maximizes the $\ell_\infty$ margin, and to Muon-Signum and Muon-Adam, which maximize a hybrid norm. Our experiments corroborate the theory and show that the identity of the margin maximized depends on the choice of optimizer. Overall, our results extend earlier lines of work on steepest descent in homogeneous models and momentum-based optimizers in linear models.


Mysterious UFO hotspots uncovered around underwater canyons off US coasts

Daily Mail - Science & tech

Horrifying next twist in the Alexander brothers case: MAUREEN CALLAHAN exposes an unthinkable perversion that's been hiding in plain sight Alexander brothers' alleged HIGH SCHOOL gang rape video: Classmates speak out on sick'taking turns' footage... as creepy unseen photos are exposed Model Cindy Crawford, 60, mocked for her'out of touch' morning routine: 'Nothing about this is normal' Kentucky mother and daughter turn down $26.5MILLION to sell their farms to secretive tech giant that wants to build data center there Tucker Carlson erupts at Trump adviser as she hurls'SLANDER' claim linking him to synagogue shooting NFL superstar Xavier Worthy spills all on Travis Kelce, the Chiefs' struggles... and having Taylor Swift as his No 1 fan Heartbreaking video shows very elderly DoorDash driver shuffle down customer's driveway with coffee order because he is too poor to retire Amber Valletta, 52, was a '90s Vogue model who made movies with Sandra Bullock and Kate Hudson, see her now Nancy Mace throws herself into Iran warzone as she goes rogue on Middle East rescue mission: 'I AM that person' Hidden toxins in kids' treats EXPOSED: Health guru Jillian Michaels' sit-down with Casey DeSantis reveals dangers lurking in popular foods READ MORE: I communicated with non-human intelligence... and what they told me proves God's existence New research has suggested that UFOs could be clustering around underwater canyons off the US coastline, raising fresh questions about whether mysterious craft could be operating beneath the ocean. An analysis of more than 80,000 reports found concentrated clusters of sightings near steep submarine canyon systems, particularly along the West Coast. The findings stem from an independent study testing the so-called'cryptoterrestrial hypothesis,' which proposes that unidentified aerial phenomena could originate from hidden non-human intelligence on Earth rather than distant planets. Using publicly available UFO sighting data and detailed ocean depth maps, the report identified correlations between reported sightings and deep underwater terrain features. The analysis also uncovered a striking geographical anomaly, with clustering patterns appearing on the West Coast but not on the East or Gulf coasts.






Face Reconstruction from Facial Templates by Learning Latent Space of a Generator Network

Neural Information Processing Systems

Among potential attacks against FR systems [Galbally et al., 2014, Biggio et al., 2015, Hadid et al., 2015, Mai et al., 2018, Marcel et al., 2023], the template inversion (TI) attack significantly jeopardizes the users' privacy. In a TI attack, the adversary gains access to templates stored in the FR system's database and aims


1d8dc55c1f6cf124af840ce1d92d1896-Paper-Conference.pdf

Neural Information Processing Systems

As inthe classical problem, weights are fixed by an adversary and elements appear in random order. In contrast to previous variants of predictions, our algorithm only has access toamuch weakerpiece ofinformation: anadditive gapc.