One of the dominant paradigms in self-supervised learning (SSL), illustrated by MoCo or DINO, aims to produce robust representations by capturing features that are insensitive to certain image transformations such as illumination, or geometric changes. This strategy is appropriate when the objective is to recognize objects independently of their appearance. However, it becomes counterproductive as soon as appearance itself constitutes the discriminative signal. In weather analysis, for example, rain streaks, snow granularity, atmospheric scattering, as well as reflections and halos, are not noise: they carry the essential information. In critical applications such as autonomous driving, ignoring these cues is risky, since grip and visibility depend directly on ground conditions and atmospheric conditions. We introduce ST-STORM, a hybrid SSL framework that treats appearance (style) as a semantic modality to be disentangled from content. Our architecture explicitly separates two latent streams, regulated by gating mechanisms. The Content branch aims at a stable semantic representation through a JEPA scheme coupled with a contrastive objective, promoting invariance to appearance variations. In parallel, the Style branch is constrained to capture appearance signatures (textures, contrasts, scattering) through feature prediction and reconstruction under an adversarial constraint. We evaluate ST-STORM on several tasks, including object classification (ImageNet-1K), fine-grained weather characterization, and melanoma detection (ISIC 2024 Challenge). The results show that the Style branch effectively isolates complex appearance phenomena (F1=97% on Multi-Weather and F1=94% on ISIC 2024 with 10% labeled data), without degrading the semantic performance (F1=80% on ImageNet-1K) of the Content branch, and improves the preservation of critical appearance

Preference-based many-objective optimization faces two obstacles: an expanding space of trade-offs and heterogeneous, context-dependent human value structures. Towards this, we propose a Bayesian framework that learns a small set of latent preference archetypes rather than assuming a single fixed utility function, modelling them as components of a Dirichlet-process mixture with uncertainty over both archetypes and their weights. To query efficiently, we designing hybrid queries that target information about (i) mode identity and (ii) within-mode trade-offs. Under mild assumptions, we provide a simple regret guarantee for the resulting mixture-aware Bayesian optimization procedure. Empirically, our method outperforms standard baselines on synthetic and real-world many-objective benchmarks, and mixture-aware diagnostics reveal structure that regret alone fails to capture.

Label noise - incorrect labels assigned to observations - can substantially degrade the performance of supervised classifiers. This paper proposes a label noise cleaning method based on Bernoulli random sampling. We show that the mean label noise levels of subsets generated by Bernoulli random sampling containing a given observation are identically distributed for all clean observations, and identically distributed, with a different distribution, for all noisy observations. Although the mean label noise levels are not independent across observations, by introducing an independent coupling we further prove that they converge to a mixture of two well-separated distributions corresponding to clean and noisy observations. By establishing a linear model between cross-validated classification errors and label noise levels, we are able to approximate this mixture distribution and thereby separate clean and noisy observations without any prior label information. The proposed method is classifier-agnostic, theoretically justified, and demonstrates strong performance on both simulated and real datasets.

Why the world's militaries are scrambling to create their own Starlink The reliable internet connections provided by Starlink offer a huge advantage on the battlefield. Starlink's satellite constellation provides a reliable internet connection to almost anywhere on Earth, conferring an advantage on the modern battlefield. But it is also run by controversial billionaire Elon Musk, presenting a risk to militaries that could easily find themselves cut off. So, now countries are racing to build their own version. The Starlink network consists of almost 10,000 satellites that offer internet connections across most of the planet via small dishes on the ground.

The US and Iran are trading blows in the Gulf with a simple drone that costs as little as $50,000 to make. But why is a slow, cheap and relatively primitive drone seeing use in 2026 alongside hypersonic missiles and stealth jets? Iran invented the relatively simple Shahed 136 attack drone, but is now fending off US copies launched against it in combat. Why, when the US military has expensive, cutting-edge and hi-tech weapons, is it making flimsy drones powered by a motorbike engine? Iranian company Shahed Aviation Industries originally designed the 136.

A clump of human brain cells can play the classic computer game . While its performance is not up to par with humans, experts say it brings biological computers a step closer to useful real-world applications, like controlling robot arms. In 2021, the Australian company Cortical Labs used its neuron-powered computer chips to play . The chips consisted of clumps of more than 800,000 living brain cells grown on top of microelectrode arrays that can both send and receive electrical signals. Researchers had to carefully train the chips to control the paddles on either side of the screen.

They are used to test the proposed enhancement approach iGA T. In general, ADP employs an ensemble by averaging, i.e., (C 1) ( C 1) Adversarial examples are generated to compute the losses by using the PGD attack. Our main theorem builds on a supporting Lemma 2.1. We start from the cross-entropy loss curvature measured by Eq. The above new expression of T (x) helps bound the difference between h(x) and h(x). Note that these three cases are mutually exclusive.

Guided by this theory, we propose an effective approach to improve ensemble adversarial defense, named interactive global adversarial training (iGA T).

Along with the dataset, two tasks are proposed to evaluate machines'