Online continual learning (OCL) methods adapt to changing environments without forgetting past knowledge. Similarly, online time series forecasting (OTSF) is a real-world problem where data evolve in time and success depends on both rapid adaptation and long-term memory. Indeed, time-varying and regime-switching forecasting models have been extensively studied, offering a strong justification for the use of OCL in these settings. Building on recent work that applies OCL to OTSF, this paper aims to strengthen the theoretical and practical connections between time series methods and OCL. First, we reframe neural network optimization as a parameter filtering problem, showing that natural gradient descent is a score-driven method and proving its information-theoretic optimality. Then, we show that using a Student's t likelihood in addition to natural gradient induces a bounded update, which improves robustness to outliers. Finally, we introduce Natural Score-driven Replay (NatSR), which combines our robust optimizer with a replay buffer and a dynamic scale heuristic that improves fast adaptation at regime drifts. Empirical results demonstrate that NatSR achieves stronger forecasting performance than more complex state-of-the-art methods.

We propose a federated learning framework for the calibration of parametric insurance indices under heterogeneous renewable energy production losses. Producers locally model their losses using Tweedie generalized linear models and private data, while a common index is learned through federated optimization without sharing raw observations. The approach accommodates heterogeneity in variance and link functions and directly minimizes a global deviance objective in a distributed setting. We implement and compare FedAvg, FedProx and FedOpt, and benchmark them against an existing approximation-based aggregation method. An empirical application to solar power production in Germany shows that federated learning recovers comparable index coefficients under moderate heterogeneity, while providing a more general and scalable framework.

Fairness-aware machine learning has recently attracted various communities to mitigate discrimination against certain societal groups in data-driven tasks. For fair supervised learning, particularly in pre-processing, there have been two main categories: data fairness and task-tailored fairness. The former directly finds an intermediate distribution among the groups, independent of the type of the downstream model, so a learned downstream classification/regression model returns similar predictive scores to individuals inputting the same covariates irrespective of their sensitive attributes. The latter explicitly takes the supervised learning task into account when constructing the pre-processing map. In this work, we study algorithmic fairness for supervised learning and argue that the data fairness approaches impose overly strong regularization from the perspective of the HGR correlation. This motivates us to devise a novel pre-processing approach tailored to supervised learning. We account for the trade-off between fairness and utility in obtaining the pre-processing map. Then we study the behavior of arbitrary downstream supervised models learned on the transformed data to find sufficient conditions to guarantee their fairness improvement and utility preservation. To our knowledge, no prior work in the branch of task-tailored methods has theoretically investigated downstream guarantees when using pre-processed data. We further evaluate our framework through comparison studies based on tabular and image data sets, showing the superiority of our framework which preserves consistent trade-offs among multiple downstream models compared to recent competing models. Particularly for computer vision data, we see our method alters only necessary semantic features related to the central machine learning task to achieve fairness.

When you purchase through links in our articles, we may earn a small commission. Most floodlight cams need hardwired power, limiting your installation options. This battery-powered model can go anywhere, and it has a solar panel, too! Single floodlight isn't as bright as you get with hardwired models Despite a couple of minor bugs, this low-cost, battery-powered floodlight camera knocks it out of the park in most respects. The "Kit" in TP-Link Tapo C615F Kit refers to the inclusion of a solar panel that comes with this full-featured security camera/floodlight combo to keep its battery charged.

China's lithium batteries aren't always "made in China." Companies like BYD and CATL are building factories on nearly every continent. THE symbolism was clear last June when Emmanuel Macron, surrounded by factory workers, held up a sleek lithium battery in his right hand and a mining lamp in his left. He was in Douai, a northern French city with a coal mining history dating back to the 1700s. The city is now also the site of a battery factory, which would allow France to produce all parts of electric vehicles domestically. This factory, Macron declared, represented an "economic and ecological revolution."

It's everything you wanted for the United States--but done better in China. China's political leaders laid out an ambitious industrial plan: By 2025, they pledged, their country would be a world capital, with the goal of moving from "Chinese speed to Chinese quality, the transformation of Chinese products to Chinese brands." This is the difference, they wrote, between "Made in China" and "Created in China." At WIRED, we never take what the government (ours or anybody else's) says at face value. Still, as journalists, we respect the ability to hit a deadline.

Thousands of Companies Are Driving China's AI Boom. How the Cyberspace Administration of China inadvertently made a guide to the country's homegrown AI revolution. When DeepSeek burst onto the global stage in January 2025, it seemed to appear out of nowhere. But the large language model was just one of the thousands of generative AI tools that have been released in China since 2023--and there's a public archive of every single one of them. Here are 23 ways China is rewiring the future .

What could possibly go wrong? The 4-foot-tall humanoid robot that's in front of me seems, quite honestly, a bit drunk. After 30 seconds or so it abruptly stops, then strides toward me with an arm outstretched. The little robot is at the World Artificial Intelligence Conference, on the banks of the Huangpu river in Shanghai. The convention center is teeming with humanoids --dancing ones, box-toting ones, robot dog-walking ones doing circuits around trade show booths. A few lie slumped in a corner as their batteries recharge. A Unitree humanoid robot modified for experimental purposes at the BAAI.

You've Never Heard of China's Greatest Sci-Fi Novel Thousands of authors. is barely known outside China--but it contains the secret to the country's modernization and malaise. Ma Qianzhu was unsatisfied with Chinese progress. An engineer at a large state-owned enterprise, he belonged to a generation that grew up believing engineering is destiny, that China's future would be built, bolt by bolt, by people like him. Then Ma discovered something extraordinary: a wormhole to the late Ming Dynasty. With more than 500 peers, he commandeered a ship and traveled back in time 400 years, to a preindustrial China wracked by foreign invasion and internal decay. Their mission: trigger an industrial revolution in the past that would, in the future, make modern China great (again).

China's Renewable Energy Revolution Is a Huge Mess That Might Save the World A global onslaught of cheap Chinese green power is upending everything in its path. No one is ready for its repercussions. There's a particular kind of sci-fi nerd who equates fusion tech with utopia. If we could only harness the engine of the stars, it would uncork near limitless energy and neatly sweep away a whole mess of humanity's problems. But how would that work exactly? What would the transition look like?