Blind source separation, particularly through independent component analysis (ICA), is widely utilized across various signal processing domains for disentangling underlying components from observed mixed signals, owing to its fully data-driven nature that minimizes reliance on prior assumptions. However, conventional ICA methods rely on an assumption of linear mixing, limiting their ability to capture complex nonlinear relationships and to maintain robustness in noisy environments. In this work, we present deep deterministic nonlinear independent component analysis (DDICA), a novel deep neural network-based framework designed to address these limitations. DDICA leverages a matrix-based entropy function to directly optimize the independence criterion via stochastic gradient descent, bypassing the need for variational approximations or adversarial schemes. This results in a streamlined training process and improved resilience to noise. We validated the effectiveness and generalizability of DDICA across a range of applications, including simulated signal mixtures, hyperspectral image unmixing, modeling of primary visual receptive fields, and resting-state functional magnetic resonance imaging (fMRI) data analysis. Experimental results demonstrate that DDICA effectively separates independent components with high accuracy across a range of applications. These findings suggest that DDICA offers a robust and versatile solution for blind source separation in diverse signal processing tasks.

Topological methods have the potential of exploring data clouds without making assumptions on their the structure. Here we propose a hierarchical topological clustering algorithm that can be implemented with any distance choice. The persistence of outliers and clusters of arbitrary shape is inferred from the resulting hierarchy. We demonstrate the potential of the algorithm on selected datasets in which outliers play relevant roles, consisting of images, medical and economic data. These methods can provide meaningful clusters in situations in which other techniques fail to do so.

Graphs are commonly used in machine learning to model relationships between instances. Consider the task of predicting the political preferences of users in a social network; to solve this task one should consider, both, the features of each individual user and the relationships between them. However, oftentimes one is not interested in the label of a single instance but rather in the distribution of labels over a set of instances; e.g., when predicting the political preferences of users, the overall prevalence of a given opinion might be of higher interest than the opinion of a specific person. This label prevalence estimation task is commonly referred to as quantification learning (QL). Current QL methods for tabular data are typically based on the so-called prior probability shift (PPS) assumption which states that the label-conditional instance distributions should remain equal across the training and test data. In the graph setting, PPS generally does not hold if the shift between training and test data is structural, i.e., if the training data comes from a different region of the graph than the test data. To address such structural shifts, an importance sampling variant of the popular adjusted count quantification approach has previously been proposed. In this work, we extend the idea of structural importance sampling to the state-of-the-art KDEy quantification approach. We show that our proposed method adapts to structural shifts and outperforms standard quantification approaches.

A first-order method for nonconvex-strongly-concave constrained minimax optimization Zhaosong Lu Sanyou Mei May 12, 2024 (Revised: October 23, 2025) Abstract In this paper we study a nonconvex-strongly-concave constrained minimax problem. Specifically, we propose a first-order augmented Lagrangian method for solving it, whose subproblems are nonconvex-strongly-concave unconstrained minimax problems and suitably solved by a first-order method developed in this paper that leverages the strong concavity structure. Under suitable assumptions, the proposed method achieves an operation complexity of O(ε 3.5 log ε 1), measured in terms of its fundamental operations, for finding an ε-KKT solution of the constrained minimax problem, which improves the previous best-known operation complexity by a factor of ε 0.5 . Keywords: minimax optimization, augmented Lagrangian method, first-order method, operation complexity Mathematics Subject Classification: 90C26, 90C30, 90C47, 90C99, 65K05 1 Introduction In this paper, we consider a nonconvex-strongly-concave constrained minimax problem F = min c(x) 0 max d(x,y) 0 {F (x,y):= f (x, y) + p(x) q(y)}. Assume that problem (1) has at least one optimal solution and the following additional assumptions hold.

The new Atlas will be deployed at Hyundai and Google DeepMind first. After years of testing its humanoid robot (and forcing it to dance), Boston Dynamics' Atlas is entering production . The robotics company says the final product version of the robot is being built now, and the first companies that will receive deployments are Hyundai, Boston Dynamics' majority shareholder, and Google DeepMind, the firm's newly minted AI partner. This final enterprise version of Atlas can perform a wide array of industrial tasks, according to Boston Dynamics, and is specifically designed with consistency and reliability in mind. Atlas can work autonomously, via a teleoperator or with a tablet steering interface, and the robot is both strong and durable.

Jensen Huang Says Nvidia's New Vera Rubin Chips Are in'Full Production' The chip giant says Vera Rubin will sharply cut the cost of training and running AI models, strengthening the appeal of its integrated computing platform. Nvidia CEO Jensen Huang says that the company's next-generation AI superchip platform, Vera Rubin, is on schedule to begin arriving to customers later this year. "Today, I can tell you that Vera Rubin is in full production," Huang said during a press event on Monday at the annual CES technology trade show in Las Vegas. Rubin will cut the cost of running AI models to about one-tenth of Nvidia's current leading chip system, Blackwell, the company told analysts and journalists during a call on Sunday. Nvidia also said Rubin can train certain large models using roughly one-fourth as many chips as Blackwell requires.

The partnership between Uber, Lucid and Nuro plans to launch its robotaxi fleet in San Francisco later this year. Waymo is getting a good look at the competition as Uber revealed the design of its robotaxi that's due to launch in San Francisco later this year. The upcoming robotaxi is a result of a partnership announced in July between Uber, Lucid and Nuro. The plan is still to deploy at least 20,000 Lucid EVs that will use the Nuro Driver autonomous driving tech and be available through the Uber platform. It's important to note that the robotaxi reveal will be a production intent design, so there may be some modifications to the version that will eventually hit the streets.

Acer's Predator Helios Neo 16S AI laptop can be outfitted with Intel's new Core Ultra 9 386H CPU The company announced the gaming computer at CES. Acer just announced the Predator Helios 16S AI gaming laptop at . This computer is filled with both bells and whistles, making it a decent choice for modern gamers. To that end, the laptop can be equipped with up to an Intel Core Ultra 9 386H processor. This is Intel's upcoming flagship mobile processor that has . The Helios 16S AI can also be outfitted with up to the NVIDIA GeForce RTX 5070 GPU.

NVIDIA has begun production on its new Vera Rubin supercomputer. NVIDIA CEO Jensen Huang presents at CES 2026, wearing a black snakeskin-like jacket. Jensen Huang took to the CES stage on Monday to share the latest from NVIDIA, and while the presentation was more a refresher of technologies the company has been working on for the past few years, there were a couple of notable announcements. NVIDIA announced Alpamayo, a family of open-source reasoning models designed to guide autonomous vehicles through difficult driving situations. The centerpiece of the release is Alpamayo 1, a 10-billion parameter chain-of-thought system NVIDIA says is capable of approaching driving more like a human being would.

Google TV's new Gemini features range from useful to unnecessary I checked out a preview of the new features during CES 2026. I met up with a few people from Google at the Encore Villas during CES (which is just 2,500 feet from my hotel but took 28 minutes to walk to, thanks to Vegas's pedestrian-averse design [also I got lost]). Once there, I saw what " more Gemini " will mean for people with a Google TV. The AI integration ranged from useful to probably unnecessary. The most useful bit, for me at least, came at the end.