This monograph presents a modular cognitive architecture for artificial intelligence grounded in the formal modeling of belief as structured semantic state. Belief states are defined as dynamic ensembles of linguistic expressions embedded within a navigable manifold, where operators enable assimilation, abstraction, nullification, memory, and introspection. Drawing from philosophy, cognitive science, and neuroscience, we develop a layered framework that enables self-regulating epistemic agents capable of reflective, goal-directed thought. At the core of this framework is the epistemic vacuum: a class of semantically inert cognitive states that serves as the conceptual origin of belief space. From this foundation, the Null Tower arises as a generative structure recursively built through internal representational capacities. The theoretical constructs are designed to be implementable in both symbolic and neural systems, including large language models, hybrid agents, and adaptive memory architectures. This work offers a foundational substrate for constructing agents that reason, remember, and regulate their beliefs in structured, interpretable ways.

Recent deep learning models for Long-term Time Series Forecasting (LTSF) often emphasize complex, handcrafted designs, while simpler architectures like linear models or MLPs have often outperformed these intricate solutions. In this paper, we revisit and organize the core ideas behind several key techniques, such as redundancy reduction and multi-scale modeling, which are frequently employed in advanced LTSF models. Our goal is to streamline these ideas for more efficient deep learning utilization. To this end, we introduce TimeCapsule, a model built around the principle of high-dimensional information compression that unifies these techniques in a generalized yet simplified framework. Specifically, we model time series as a 3D tensor, incorporating temporal, variate, and level dimensions, and leverage mode production to capture multi-mode dependencies while achieving dimensionality compression. We propose an internal forecast within the compressed representation domain, supported by the Joint-Embedding Predictive Architecture (JEPA), to monitor the learning of predictive representations. Extensive experiments on challenging benchmarks demonstrate the versatility of our method, showing that TimeCapsule can achieve state-of-the-art performance.

Donald Trump signed a series of executive orders on Friday intended to spur a "nuclear energy renaissance" through the construction of new reactors he said would satisfy the electricity demands of data centers for artificial intelligence and other emerging industries. The orders represented the president's latest foray into the policy underlying America's electricity supply. Trump declared a national energy emergency on his first day in office over and moved to undo a ban implemented by Joe Biden on new natural gas export terminals and expand oil and gas drilling in Alaska. Nuclear does not carry oil and gas's carbon emissions, but produces radioactive waste that the United States lacks a facility to permanently store. Some environmental groups have safety concerns over the reactors and their supply chain. Trump signed four orders intended to speed up the approval of nuclear reactors for defense and AI purposes, reform the Nuclear Regulatory Commission with the goal of quadrupling production of electricity over the next 25 years, revamp the regulatory process to have three experimental reactors operating by 4 July 2026 and boost investment in the technology's industrial base.

The success of unsupervised learning raises the question of whether also supervised models can be trained without using the information in the output $y$. In this paper, we demonstrate that this is indeed possible. The key step is to formulate the model as a smoother, i.e. on the form $\hat{f}=Sy$, and to construct the smoother matrix $S$ independently of $y$, e.g. by training on random labels. We present a simple model selection criterion based on the distribution of the out-of-sample predictions and show that, in contrast to cross-validation, this criterion can be used also without access to $y$. We demonstrate on real and synthetic data that $y$-free trained versions of linear and kernel ridge regression, smoothing splines, and neural networks perform similarly to their standard, $y$-based, versions and, most importantly, significantly better than random guessing.

Contextual online decision-making problems with constraints appear in a wide range of real-world applications, such as adaptive experimental design under safety constraints, personalized recommendation with resource limits, and dynamic pricing under fairness requirements. In this paper, we investigate a general formulation of sequential decision-making with stage-wise feasibility constraints, where at each round, the learner must select an action based on observed context while ensuring that a problem-specific feasibility criterion is satisfied. We propose a unified algorithmic framework that captures many existing constrained learning problems, including constrained bandits, active learning with label budgets, online hypothesis testing with Type I error control, and model calibration. Central to our approach is the concept of upper counterfactual confidence bounds, which enables the design of practically efficient online algorithms with strong theoretical guarantees using any offline conditional density estimation oracle. To handle feasibility constraints in complex environments, we introduce a generalized notion of the eluder dimension, extending it from the classical setting based on square loss to a broader class of metric-like probability divergences. This allows us to capture the complexity of various density function classes and characterize the utility regret incurred due to feasibility constraint uncertainty. Our result offers a principled foundation for constrained sequential decision-making in both theory and practice.

We investigate scaling laws for stochastic momentum algorithms with small batch on the power law random features model, parameterized by data complexity, target complexity, and model size. When trained with a stochastic momentum algorithm, our analysis reveals four distinct loss curve shapes determined by varying data-target complexities. While traditional stochastic gradient descent with momentum (SGD-M) yields identical scaling law exponents to SGD, dimension-adapted Nesterov acceleration (DANA) improves these exponents by scaling momentum hyperparameters based on model size and data complexity. This outscaling phenomenon, which also improves compute-optimal scaling behavior, is achieved by DANA across a broad range of data and target complexities, while traditional methods fall short. Extensive experiments on high-dimensional synthetic quadratics validate our theoretical predictions and large-scale text experiments with LSTMs show DANA's improved loss exponents over SGD hold in a practical setting.

The Madden-Julian Oscillation, or MJO, is a significant weather pattern that affects weather, influencing rainfall, temperature, and even storm frequency and intensity. When the MJO is active, it can affect the weather globally. To better predict weather changes with 3-4 weeks in advance, we rely on the ability to predict the MJO's activity. Data-driven methods such as the ones that rely on deep neural networks have been recently employed to make such predictions. By examining existing MJO patterns, neural networks attempt to predict upcoming ones. However, while neural networks are robust enough to predict the MJO's activity, they do not provide confidence intervals for those predictions. To address this shortcoming, we use a model known as the "Gaussian process" or GP. This statistical tool is distinctive because it not only provides predictions but also quantifies the level of confidence in them.

Personal use of this material is permitted. Abstract --Virtual Platforms (VPs) enable early software validation of autonomous systems' electronics, reducing costs and time-to-market. While many VPs support both functional and non-functional simulation (e.g., timing, power), they lack the capability of simulating the environment in which the system operates. In contrast, robotics simulators lack accurate timing and power features. This twofold shortcoming limits the effectiveness of the design flow, as the designer can not fully evaluate the features of the solution under development. This paper presents a novel, fully open-source framework bridging this gap by integrating a robotics simulator (Webots) with a VP for RISC-V-based systems (MESSY). The framework enables a holistic, mission-level, energy-aware co-simulation of electronics in their surrounding environment, streamlining the exploration of design configurations and advanced power management policies. Virtual Platforms (VPs) enable comprehensive system modeling and simulation before physical production [1] and are thus a crucial resource in the design of modern embedded systems, characterized by heterogeneity and tight integration with the physical environment.

Electrostatic adhesion is widely used in mobile robotics, haptics, and robotic end effectors for its adaptability to diverse substrates and low energy consumption. Force sensing is important for feedback control, interaction, and monitoring in the EA system. However, EA force monitoring often relies on bulky and expensive sensors, increasing the complexity and weight of the entire system. This paper presents an acoustic-pressure-based method to monitor EA forces without contacting the adhesion pad. When the EA pad is driven by a bipolar square-wave voltage to adhere a conductive object, periodic acoustic pulses arise from the EA system. We employed a microphone to capture these acoustic pressure signals and investigate the influence of peak pressure values. Results show that the peak value of acoustic pressure increased with the mass and contact area of the adhered object, as well as with the amplitude and frequency of the driving voltage. We applied this technique to mass estimation of various objects and simultaneous monitoring of two EA systems. Then, we integrated this technique into an EA end effector that enables monitoring the change of adhered object mass during transport. The proposed technique offers a low-cost, non-contact, and multi-object monitoring solution for EA end effectors in handling tasks.

--The growing complexity of power systems has made accurate load forecasting more important than ever . An increasing number of advanced load forecasting methods have been developed. However, the static design of current methods offers no mechanism for human-model interaction. As the primary users of forecasting models, system operators often find it difficult to understand and apply these advanced models, which typically requires expertise in artificial intelligence (AI). This also prevents them from incorporating their experience and real-world contextual understanding into the forecasting process. Recent breakthroughs in large language models (LLMs) offer a new opportunity to address this issue. By leveraging their natural language understanding and reasoning capabilities, we propose an LLM-based multi-agent collaboration framework to bridge the gap between human operators and forecasting models. A set of specialized agents is designed to perform different tasks in the forecasting workflow and collaborate via a dedicated communication mechanism. Our experiments demonstrate that the interactive load forecasting accuracy can be significantly improved when users provide proper insight in key stages. Our cost analysis shows that the framework remains affordable, making it practical for real-world deployment. With the boom of artificial intelligence, a wide range of forecasting algorithms have been proposed recently, many of which have demonstrated impressive performance. However, these forecasting methods become static once designed, offering no mechanism for interaction between the model and human users. This lack of interaction creates major barriers to the practical use of the forecasting methods.