In recent years, the Shapley value and SHAP explanations have emerged as one of the most dominant paradigms for providing post-hoc explanations of black-box models. Despite their well-founded theoretical properties, many recent works have focused on the limitations in both their computational efficiency and their representation power. The underlying connection with additive models, however, is left critically under-emphasized in the current literature. In this work, we find that a variational perspective linking GAM models and SHAP explanations is able to provide deep insights into nearly all recent developments. In light of this connection, we borrow in the other direction to develop a new method to train interpretable GAM models which are automatically purified to compute the Shapley value in a single forward pass. Finally, we provide theoretical results showing the limited representation power of GAM models is the same Achilles' heel existing in SHAP and discuss the implications for SHAP's modern usage in CV and NLP.

In many organisations, accurate forecasts are essential for making informed decisions for a variety of applications from inventory management to staffing optimization. Whatever forecasting model is used, changes in the underlying process can lead to inaccurate forecasts, which will be damaging to decision-making. At the same time, models are becoming increasingly complex and identifying change through direct modelling is problematic. We present a novel framework for online monitoring of forecasts to ensure they remain accurate. By utilizing sequential changepoint techniques on the forecast errors, our framework allows for the real-time identification of potential changes in the process caused by various external factors. We show theoretically that some common changes in the underlying process will manifest in the forecast errors and can be identified faster by identifying shifts in the forecast errors than within the original modelling framework. Moreover, we demonstrate the effectiveness of this framework on numerous forecasting approaches through simulations and show its effectiveness over alternative approaches. Finally, we present two concrete examples, one from Royal Mail parcel delivery volumes and one from NHS A\&E admissions relating to gallstones.

Conformal prediction provides a powerful framework for constructing prediction intervals with finite-sample guarantees, yet its robustness under distribution shifts remains a significant challenge. This paper addresses this limitation by modeling distribution shifts using L\'evy-Prokhorov (LP) ambiguity sets, which capture both local and global perturbations. We provide a self-contained overview of LP ambiguity sets and their connections to popular metrics such as Wasserstein and Total Variation. We show that the link between conformal prediction and LP ambiguity sets is a natural one: by propagating the LP ambiguity set through the scoring function, we reduce complex high-dimensional distribution shifts to manageable one-dimensional distribution shifts, enabling exact quantification of worst-case quantiles and coverage. Building on this analysis, we construct robust conformal prediction intervals that remain valid under distribution shifts, explicitly linking LP parameters to interval width and confidence levels. Experimental results on real-world datasets demonstrate the effectiveness of the proposed approach.

It has recently been argued that AI models' representations are becoming aligned as their scale and performance increase. Empirical analyses have been designed to support this idea and conjecture the possible alignment of different representations toward a shared statistical model of reality. In this paper, we propose a learning-theoretic perspective to representation alignment. First, we review and connect different notions of alignment based on metric, probabilistic, and spectral ideas. Then, we focus on stitching, a particular approach to understanding the interplay between different representations in the context of a task. Our main contribution here is relating properties of stitching to the kernel alignment of the underlying representation. Our results can be seen as a first step toward casting representation alignment as a learning-theoretic problem.

A substantial body of work in machine learning (ML) and randomized numerical linear algebra (RandNLA) has exploited various sorts of random sketching methodologies, including random sampling and random projection, with much of the analysis using Johnson--Lindenstrauss and subspace embedding techniques. Recent studies have identified the issue of inversion bias -- the phenomenon that inverses of random sketches are not unbiased, despite the unbiasedness of the sketches themselves. This bias presents challenges for the use of random sketches in various ML pipelines, such as fast stochastic optimization, scalable statistical estimators, and distributed optimization. In the context of random projection, the inversion bias can be easily corrected for dense Gaussian projections (which are, however, too expensive for many applications). Recent work has shown how the inversion bias can be corrected for sparse sub-gaussian projections. In this paper, we show how the inversion bias can be corrected for random sampling methods, both uniform and non-uniform leverage-based, as well as for structured random projections, including those based on the Hadamard transform. Using these results, we establish problem-independent local convergence rates for sub-sampled Newton methods.

Elon Musk's artificial intelligence startup xAI has introduced Grok-3, the latest iteration of its chatbot that integrates with X, formerly Twitter. Grok-3 debut comes at a critical moment in the AI arms race as Musk looks to compete with the Chinese AI firm DeepSeek, Microsoft-backed OpenAI and Google. Musk's bot has seen less widespread adoption than DeepSeek's namesake chatbot, which wowed the world weeks ago and caused panic in stock markets, as well as OpenAI's ChatGPT and Google's Gemini. Grok-3 is being rolled out immediately to Premium subscribers of X, the social media platform owned by Musk. The chatbot can generate texts and images without many of the common guardrails against sexually suggestive imagery, vulgarity or the reproduction of well-known people's likenesses. "Grok-3 across the board is in a league of its own," Musk said during a livestream alongside three xAI engineers late on Monday.

Kurt Knutsson joins "Fox & Friends" to discuss bank scams and a self-driving car that trapped a rider inside. Tech support scams have been around for years, but a new variant called the Phantom Hacker scam is rapidly gaining traction. It has cost victims, primarily older Americans, over 500 million since 2023. This scam is particularly deceptive because it unfolds in three carefully orchestrated phases and uses AI-powered social engineering tactics to avoid detection. Attackers leverage caller ID spoofing and AI-generated voices to make their scheme more persuasive, but there are ways to protect yourself.

New deep fakes are all over the internet -- and you won't believe the new ones Raymond Arroyo has located. The trial has begun of a Democratic political consultant who has admitted to sending artificial intelligence (AI) generated robocalls mimicking President Biden ahead of the 2024 New Hampshire primary. Steve Kramer faces a 6 million fine and more than two dozen criminal charges after he hired a magician to create a deepfake of President Biden urging New Hampshire voters not to participate in the primary. The fines, proposed by the Federal Communications Commission (FCC), are the first involving AI technology. Former president Joe Biden speaks on the phone during a National Small Business Week event in the Rose Garden of the White House in Washington, DC, on May 1, 2023, left.

Ukraine and Russia are now three years into what has been called the first drone war: not the first in which they were used, but the first in which they have been a major factor on the battlefield. What lessons have others drawn about the shape of future wars? "Drones are here to stay, and they will be everywhere – on the ground, in the air and at sea – in numbers," says Oleksandra Molloy at the University of New South Wales in Canberra, Australia. "The point of no return was…

President Xi Jinping's embrace of Chinese tech bosses in a rare public meeting is fueling hope Beijing is shifting its stance to give the private sector a freer hand as it fights a trade war with U.S. President Donald Trump. Four years after launching a regulatory crackdown that plunged the tech sector into turmoil, China's top leader sat down publicly for the first time with Alibaba co-founder Jack Ma, whose firm bore the brunt of that campaign. Also on the guest list Monday were rising stars from robotics start-up Unitree, electric car giant BYD and AI newcomer DeepSeek -- firms rolling out world-beating innovations despite U.S. export controls. While a similar show of support from Xi in 2018 proved fleeting, developing national tech champions is core to Beijing's plan for boosting the economy as it deflates a bubble in the property market that once drove about a quarter of growth. Underscoring the importance of spurring innovation, high-tech industries contributed to 15% of gross domestic product last year and are set to overtake the housing sector in 2026, according to Bloomberg Economics.