Exploding energy costs have been blamed for the closure of more than 60 public swimming pools across Britain over the past four years. And with the bills for some expected to rise by £100,000 this year, it has left leisure centres scrabbling around for ways to keep the facilities running. It may sound far-fetched, but one leisure centre in Devon is using computer power to heat its swimming pool. The idea works by placing 12 computers inside a white box which is then surrounded by oil to capture the waste heat they produce -- in a similar way to another concept that uses computer servers to heat water in people's homes. Innovative: It may sound far-fetched, but Exmouth Leisure Centre in Devon is using computer power to heat its swimming pool.

Public swimming pools facing closure because of soaring energy bills have been offered a lifeline via new technology to heat the water. Mark Bjornsgaard, the chief executive of the tech startup Deep Green, has trialled the idea in Exmouth, Devon. He has put a small computer data processing centre underneath the pool and the energy from it heats the water. The idea has taken off and up to 20 public pools could be upgraded to the heat system this year. "We built a small data centre in Exmouth leisure centre. Most normal data centres waste the heat that the computers generate. We capture ours and we give it for free to the swimming pool to heat the pool," Bjornsgaard told BBC Radio 4's Today programme.

How does an iPhone predict the next word you're going to type in your messages? The technology behind this, and also at the core of many AI applications, is called a transformer; a deep-learning algorithm that detects patterns in datasets. Now, researchers at EPFL and KAIST have created a transformer for Metal-Organic Frameworks (MOFs), a class of porous crystalline materials. By combining organic linkers with metal nodes, chemists can synthesize millions of different materials with potential applications in energy storage and gas separation. The "MOFtransformer" is designed to be the ChatGPT for researchers that study MOFs.

Abstract-- Achieving reactive robot behavior in complex dynamic environments is still challenging as it relies on being able to solve trajectory optimization problems quickly enough, such that we can replan the future motion at frequencies which are sufficiently high for the task at hand. We argue that current limitations in Model Predictive Control (MPC) for robot manipulators arise from inefficient, high-dimensional trajectory representations and the negligence of time-optimality in the trajectory optimization process. Therefore, we propose a motion optimization framework that optimizes jointly over space and time, generating smooth and timing-optimal robot trajectories in joint-space. Such task settings require performance. Compared to gradient-based optimization, the robot to be reactive to unforeseen changes in stochastic approaches typically also achieve higher robustness the environment, e.g., due to dynamic obstacles, as well to difficult reward landscapes due to their exploratory as to be robust and compliant when operating alongside properties [5].

Abstract-- Soft robotic grasping has rapidly spread through the academic robotics community in recent years and pushed into industrial applications. At the same time, multimaterial 3D printing has become widely available, enabling the monolithic manufacture of devices containing rigid and elastic sections. We propose a novel design technique that leverages both technologies and can automatically design bespoke soft robotic grippers for fruit-picking and similar applications. We demonstrate the novel topology optimisation formulation that generates multi-material soft grippers, can solve internal and external pressure boundaries, and investigate methods to produce air-tight designs. Compared to existing methods, it vastly expands the searchable design space while increasing simulation accuracy.

This paper extends the gap-based navigation technique in Potential Gap by guaranteeing safety for nonholonomic robots for all tiers of the local planner hierarchy, so called Safer Gap. The first tier generates a Bezier-based collision-free path through gaps. A subset of navigable free-space from the robot through a gap, called the keyhole, is defined to be the union of the largest collision-free disc centered on the robot and a trapezoidal region directed through the gap. It is encoded by a shallow neural network zeroing barrier function (ZBF). Nonlinear model predictive control (NMPC), with Keyhole ZBF constraints and output tracking of the Bezier path, synthesizes a safe kinematically-feasible trajectory. Low-level use of the Keyhole ZBF within a point-wise optimization-based safe control synthesis module serves as a final safety layer. Simulation and experimental validation of Safer Gap confirm its collision-free navigation properties.

With the rapid advancement of 5G networks, billions of smart Internet of Things (IoT) devices along with an enormous amount of data are generated at the network edge. While still at an early age, it is expected that the evolving 6G network will adopt advanced artificial intelligence (AI) technologies to collect, transmit, and learn this valuable data for innovative applications and intelligent services. However, traditional machine learning (ML) approaches require centralizing the training data in the data center or cloud, raising serious user-privacy concerns. Federated learning, as an emerging distributed AI paradigm with privacy-preserving nature, is anticipated to be a key enabler for achieving ubiquitous AI in 6G networks. However, there are several system and statistical heterogeneity challenges for effective and efficient FL implementation in 6G networks. In this article, we investigate the optimization approaches that can effectively address the challenging heterogeneity issues from three aspects: incentive mechanism design, network resource management, and personalized model optimization. We also present some open problems and promising directions for future research.

The advent of AlphaGo and its successors marked the beginning of a new paradigm in playing games using artificial intelligence. This was achieved by combining Monte Carlo tree search, a planning procedure, and deep learning. While the impact on the domain of games has been undeniable, it is less clear how useful similar approaches are in applications beyond games and how they need to be adapted from the original methodology. We review 129 peer-reviewed articles detailing the application of neural Monte Carlo tree search methods in domains other than games. Our goal is to systematically assess how such methods are structured in practice and if their success can be extended to other domains. We find applications in a variety of domains, many distinct ways of guiding the tree search using learned policy and value functions, and various training methods. Our review maps the current landscape of algorithms in the family of neural monte carlo tree search as they are applied to practical problems, which is a first step towards a more principled way of designing such algorithms for specific problems and their requirements.

Printed electronics (PE) promises on-demand fabrication, low non-recurring engineering costs, and sub-cent fabrication costs. It also allows for high customization that would be infeasible in silicon, and bespoke architectures prevail to improve the efficiency of emerging PE machine learning (ML) applications. Nevertheless, large feature sizes in PE prohibit the realization of complex ML models in PE, even with bespoke architectures. In this work, we present an automated, cross-layer approximation framework tailored to bespoke architectures that enable complex ML models, such as Multi-Layer Perceptrons (MLPs) and Support Vector Machines (SVMs), in PE. Our framework adopts cooperatively a hardware-driven coefficient approximation of the ML model at algorithmic level, a netlist pruning at logic level, and a voltage over-scaling at the circuit level. Extensive experimental evaluation on 12 MLPs and 12 SVMs and more than 6000 approximate and exact designs demonstrates that our model-to-circuit cross-approximation delivers power and area optimal designs that, compared to the state-of-the-art exact designs, feature on average 51% and 66% area and power reduction, respectively, for less than 5% accuracy loss. Finally, we demonstrate that our framework enables 80% of the examined classifiers to be battery-powered with almost identical accuracy with the exact designs, paving thus the way towards smart complex printed applications.

Neural-network quantum molecular dynamics (NNQMD) simulations based on machine learning are revolutionizing atomistic simulations of materials by providing quantum-mechanical accuracy but orders-of-magnitude faster, illustrated by ACM Gordon Bell prize (2020) and finalist (2021). State-of-the-art (SOTA) NNQMD model founded on group theory featuring rotational equivariance and local descriptors has provided much higher accuracy and speed than those models, thus named Allegro (meaning fast). On massively parallel supercomputers, however, it suffers a fidelity-scaling problem, where growing number of unphysical predictions of interatomic forces prohibits simulations involving larger numbers of atoms for longer times. Here, we solve this problem by combining the Allegro model with sharpness aware minimization (SAM) for enhancing the robustness of model through improved smoothness of the loss landscape. The resulting Allegro-Legato (meaning fast and "smooth") model was shown to elongate the time-to-failure $t_\textrm{failure}$, without sacrificing computational speed or accuracy. Specifically, Allegro-Legato exhibits much weaker dependence of timei-to-failure on the problem size, $t_{\textrm{failure}} \propto N^{-0.14}$ ($N$ is the number of atoms) compared to the SOTA Allegro model $\left(t_{\textrm{failure}} \propto N^{-0.29}\right)$, i.e., systematically delayed time-to-failure, thus allowing much larger and longer NNQMD simulations without failure. The model also exhibits excellent computational scalability and GPU acceleration on the Polaris supercomputer at Argonne Leadership Computing Facility. Such scalable, accurate, fast and robust NNQMD models will likely find broad applications in NNQMD simulations on emerging exaflop/s computers, with a specific example of accounting for nuclear quantum effects in the dynamics of ammonia.