Recent developments in Machine Learning approaches for modelling physical systems have begun to mirror the past development of numerical methods in the computational sciences. In this survey, we begin by providing an example of this with the parallels between the development trajectories of graph neural network acceleration for physical simulations and particle-based approaches. We then give an overview of simulation approaches, which have not yet found their way into state-of-the-art Machine Learning methods and hold the potential to make Machine Learning approaches more accurate and more efficient. We conclude by presenting an outlook on the potential of these approaches for making Machine Learning models for science more efficient.

We contribute to the vastly growing field of machine learning for engineering systems by demonstrating that equivariant graph neural networks have the potential to learn more accurate dynamic-interaction models than their non-equivariant counterparts. We benchmark two well-studied fluid flow systems, namely the 3D decaying Taylor-Green vortex and the 3D reverse Poiseuille flow, and compare equivariant graph neural networks to their non-equivariant counterparts on different performance measures, such as kinetic energy or Sinkhorn distance. Such measures are typically used in engineering to validate numerical solvers. Our main findings are that while being rather slow to train and evaluate, equivariant models learn more physically accurate interactions. This indicates opportunities for future work towards coarse-grained models for turbulent flows, and generalization across system dynamics and parameters.

Large language models (LLMs) have significantly transformed the landscape of artificial intelligence by demonstrating their ability in generating human-like text across diverse topics. However, despite their impressive capabilities, LLMs lack recent information and often employ imprecise language, which can be detrimental in domains where accuracy is crucial, such as climate change. In this study, we make use of recent ideas to harness the potential of LLMs by viewing them as agents that access multiple sources, including databases containing recent and precise information about organizations, institutions, and companies. We demonstrate the effectiveness of our method through a prototype agent that retrieves emission data from ClimateWatch (https://www.climatewatchdata.org/) and leverages general Google search. By integrating these resources with LLMs, our approach overcomes the limitations associated with imprecise language and delivers more reliable and accurate information in the critical domain of climate change. This work paves the way for future advancements in LLMs and their application in domains where precision is of paramount importance.

In recent years, the robotics community has witnessed the development of several software stacks for ground and articulated robots, such as Navigation2 and MoveIt. However, the same level of collaboration and standardization is yet to be achieved in the field of aerial robotics, where each research group has developed their own frameworks. This work presents Aerostack2, a framework for the development of autonomous aerial robotics systems that aims to address the lack of standardization and fragmentation of efforts in the field. Built on ROS 2 middleware and featuring an efficient modular software architecture and multi-robot orientation, Aerostack2 is a versatile and platform-independent environment that covers a wide range of robot capabilities for autonomous operation. Its major contributions include providing a logical level for specifying missions, reusing components and sub-systems for aerial robotics, and enabling the development of complete control architectures. All major contributions have been tested in simulation and real flights with multiple heterogeneous swarms. Aerostack2 is open source and community oriented, democratizing the access to its technology by autonomous drone systems developers.

As geospatial machine learning models and maps derived from their predictions are increasingly used for downstream analyses in science and policy, it is imperative to evaluate their accuracy and applicability. Geospatial machine learning has key distinctions from other learning paradigms, and as such, the correct way to measure performance of spatial machine learning outputs has been a topic of debate. In this paper, I delineate unique challenges of model evaluation for geospatial machine learning with global or remotely sensed datasets, culminating in concrete takeaways to improve evaluations of geospatial model performance. Geospatial machine learning (ML), for example with remotely sensed data, is being used across consequential domains, including public health (Nilsen et al., 2021; Draidi Areed et al., 2022) conservation (Sofaer et al., 2019), food security (Nakalembe, 2018), and wealth estimation (Jean et al., 2016; Chi et al., 2022). By both their use and their very nature, geospatial predictions have a purpose beyond model benchmarking; mapped data are to be read, scrutinized, and acted upon.

This paper presents a review of the field of reinforcement learning (RL), with a focus on providing a comprehensive overview of the key concepts, techniques, and algorithms for beginners. RL has a unique setting, jargon, and mathematics that can be intimidating for those new to the field or artificial intelligence more broadly. While many papers review RL in the context of specific applications, such as games, healthcare, finance, or robotics, these papers can be difficult for beginners to follow due to the inclusion of non-RL-related work and the use of algorithms customized to those specific applications. To address these challenges, this paper provides a clear and concise overview of the fundamental principles of RL and covers the different types of RL algorithms. For each algorithm/method, we outline the main motivation behind its development, its inner workings, and its limitations. The presentation of the paper is aligned with the historical progress of the field, from the early 1980s Q-learning algorithm to the current state-of-the-art algorithms such as TD3, PPO, and offline RL. Overall, this paper aims to serve as a valuable resource for beginners looking to construct a solid understanding of the fundamentals of RL and be aware of the historical progress of the field. It is intended to be a go-to reference for those interested in learning about RL without being distracted by the details of specific applications.

This paper addresses the challenge of efficiently solving the optimal power flow problem in real-time electricity markets. The proposed solution, named Physics-Informed Market-Aware Active Set learning OPF (PIMA-AS-OPF), leverages physical constraints and market properties to ensure physical and economic feasibility of market-clearing outcomes. Specifically, PIMA-AS-OPF employs the active set learning technique and expands its capabilities to account for curtailment in load or renewable power generation, which is a common challenge in real-world power systems. The core of PIMA-AS-OPF is a fully-connected neural network that takes the net load and the system topology as input. The outputs of this neural network include active constraints such as saturated generators and transmission lines, as well as non-zero load shedding and wind curtailments. These outputs allow for reducing the original market-clearing optimization to a system of linear equations, which can be solved efficiently and yield both the dispatch decisions and the locational marginal prices (LMPs). The dispatch decisions and LMPs are then tested for their feasibility with respect to the requirements for efficient market-clearing results. The accuracy and scalability of the proposed method is tested on a realistic 1814-bus NYISO system with current and future renewable energy penetration levels.

Researchers at America's Pacific Northwest National Laboratory (PNNL) are developing machine learning techniques to help the Feds crack down on potentially rogue nuclear weapons. Suffice to say, it's generally illegal for any individual or group to own a nuclear weapon, certainly in the United States. Yes, there are the five officially recognized nuclear-armed nations – France, Russia, China, the UK, and the US – whose governments have a stash of these devices. And there are countries that have signed the United Nations' Treaty on the Prohibition of Nuclear Weapons, meaning they've promised not to "develop, test, produce, acquire, possess, stockpile, use or threaten to use" these gadgets. So if anyone has a nuke in their possession, it's because they are a country in the official nuclear-armed club, they are a government that's produced its own nukes, a terrorist who stole, bought, or somehow built one themselves, or some other sketchy scenario, in America's eyes at least.

This open letter signed by many AI experts may seem at first glance very commendable. But if we take a closer look at this initiative, is this open letter really relevant to move towards a responsible AI? First of all, this open letter has been published by a non-profit association "aiming to steer transformative technology towards the benefit of life and away from large-scale extreme risks". Again, obviously at first glance, this organization seems to be moving in the direction of the public good. But still by digging a little… this one was financed from the beginning by our favorite blue bird guy – everyone will recognize him – who is still one of the influential advisors of this association and who also signed this open letter. A very nice man who is recognized by everyone to be of an unstoppable ethics concerning AI, isn't it?

Apple has set the dates for WWDC 2023, which will run between June 5th and June 9th. It's still an online-only affair, but there will be a "special experience" at Apple Park on the 5th for developers and students. While we expect to see software-centric upgrades, with iOS, macOS and the rest, this could also be when Apple finally debuts its mixed-reality headset. Rumors suggest it could be called Reality Pro or Reality One, and it's believed to be a standalone device with an M2 chip, dual 4K displays, advanced body tracking and controller-free input. It could be a pricey piece of hardware, even by Apple's standards, with some reports suggesting it'll cost $3,000. Get our daily audio briefings, Monday through Friday, by subscribing right here.