The interplay between models and observations is a cornerstone of the scientific method, aiming to inform which theoretical models are reflected in the observed data. Within cosmology, as both models and observations have substantially increased in complexity over time, the tools needed to enable a rigorous comparison have required updating as well. With an eye towards the next decade in cosmology, the vast data volumes to be delivered by ongoing and upcoming surveys, as well as the ever-expanding theoretical search-space, motivate a re-thinking of the statistical machinery used. In particular, we are now at a crucial juncture where we may be limited by the statistical and data-driven tools themselves rather than the quality or volume of the available data. Methods based on artificial intelligence (AI) and machine learning (ML) have recently emerged as promising tools for cosmological applications, demonstrating the ability to overcome some of the computational bottlenecks associated with traditional statistical techniques. Machine learning is starting to see increased adoption across different subfields of and for various applications within cosmology. At the same time, the nascent and emergent nature of practical artificial intelligence motivates careful continued development and significant care when it comes to their application in the sciences, as well as cognizance of their potential for broader societal impact. In this white paper, we provide an overview of some of the ways machine learning methods are becoming increasingly central to the way cosmological data is collected, analyzed, and interpreted. Along the way, we highlight our vision for necessary developments, framing these as recommendations--both technological as well as sociological--for the widespread safe and equitable adoption of machine learning methods within cosmology in the coming decade.

Federated Leaning is an emerging approach to manage cooperation between a group of agents for the solution of Machine Learning tasks, with the goal of improving each agent's performance without disclosing any data. In this paper we present a novel algorithmic architecture that tackle this problem in the particular case of Anomaly Detection (or classification or rare events), a setting where typical applications often comprise data with sensible information, but where the scarcity of anomalous examples encourages collaboration. We show how Random Forests can be used as a tool for the development of accurate classifiers with an effective insight-sharing mechanism that does not break the data integrity. Moreover, we explain how the new architecture can be readily integrated in a blockchain infrastructure to ensure the verifiable and auditable execution of the algorithm. Furthermore, we discuss how this work may set the basis for a more general approach for the design of federated ensemble-learning methods beyond the specific task and architecture discussed in this paper.

Medical images are a rich source of invaluable necessary information used by clinicians. Recent technologies have introduced many advancements for exploiting the most of this information and use it to generate better analysis. Deep learning (DL) techniques have been empowered in medical images analysis using computer-assisted imaging contexts and presenting a lot of solutions and improvements while analyzing these images by radiologists and other specialists. In this paper, we present a survey of DL techniques used for variety of tasks along with the different medical image's modalities to provide critical review of the recent developments in this direction. We have organized our paper to provide significant contribution of deep leaning traits and learn its concepts, which is in turn helpful for non-expert in medical society.

In the past decade, we have witnessed the rise of deep learning to dominate the field of artificial intelligence. Advances in artificial neural networks alongside corresponding advances in hardware accelerators with large memory capacity, together with the availability of large datasets enabled practitioners to train and deploy sophisticated neural network models that achieve state-of-the-art performance on tasks across several fields spanning computer vision, natural language processing, and reinforcement learning. However, as these neural networks become bigger, more complex, and more widely used, fundamental problems with current deep learning models become more apparent. State-of-the-art deep learning models are known to suffer from issues that range from poor robustness, inability to adapt to novel task settings, to requiring rigid and inflexible configuration assumptions. Collective behavior, commonly observed in nature, tends to produce systems that are robust, adaptable, and have less rigid assumptions about the environment configuration. Collective intelligence, as a field, studies the group intelligence that emerges from the interactions of many individuals. Within this field, ideas such as self-organization, emergent behavior, swarm optimization, and cellular automata were developed to model and explain complex systems. It is therefore natural to see these ideas incorporated into newer deep learning methods. In this review, we will provide a historical context of neural network research's involvement with complex systems, and highlight several active areas in modern deep learning research that incorporate the principles of collective intelligence to advance its current capabilities. We hope this review can serve as a bridge between the complex systems and deep learning communities.

If there's an extension I'd like to see made this survey, is a more in-depth inclusion of recent multimodal works like relying on prompting such as Multimodal Few-Shot Learning with Frozen Language Models³, which we've highlighted in a previous blog post.

Smart Cities sensing is an emerging paradigm to facilitate the transition into smart city services. The advent of the Internet of Things (IoT) and the widespread use of mobile devices with computing and sensing capabilities has motivated applications that require data acquisition at a societal scale. These valuable data can be leveraged to train advanced Artificial Intelligence (AI) models that serve various smart services that benefit society in all aspects. Despite their effectiveness, legacy data acquisition models backed with centralized Machine Learning models entail security and privacy concerns, and lead to less participation in large-scale sensing and data provision for smart city services. To overcome these challenges, Federated Learning is a novel concept that can serve as a solution to the privacy and security issues encountered within the process of data collection. This survey article presents an overview of smart city sensing and its current challenges followed by the potential of Federated Learning in addressing those challenges. A comprehensive discussion of the state-of-the-art methods for Federated Learning is provided along with an in-depth discussion on the applicability of Federated Learning in smart city sensing; clear insights on open issues, challenges, and opportunities in this field are provided as guidance for the researchers studying this subject matter.

We present a scalable tree search planning algorithm for large multi-agent sequential decision problems that require dynamic collaboration. Teams of agents need to coordinate decisions in many domains, but naive approaches fail due to the exponential growth of the joint action space with the number of agents. We circumvent this complexity through an approach that allows us to trade computation for approximation quality and dynamically coordinate actions. Our algorithm comprises three elements: online planning with Monte Carlo Tree Search (MCTS), factored representations of local agent interactions with coordination graphs, and the iterative Max-Plus method for joint action selection. We evaluate our approach on the benchmark SysAdmin domain with static coordination graphs and achieve comparable performance with much lower computation cost than our MCTS baselines. We also introduce a multi-drone delivery domain with dynamic coordination graphs, and demonstrate how our approach scales to large problems on this domain that are intractable for other MCTS methods.

There has been a lot of buzz surrounding the adoption of artificial intelligence. According to a recent report from McKinsey 57 percent of companies are now using AI in at least one function. But how much is hype and how much is built on a sound commercial base? We spoke to Mike Loukides, VP of emerging tech content at O'Reilly Media and author of O'Reilly Media's widely-cited AI Adoption in the Enterprise report, to discuss the current state of AI and what lies ahead. BN: Are we moving beyond the adoption of AI because it's new and cool to having a serious business case?

Linear-time temporal logic on finite traces (LTLf) is rapidly becoming a de-facto standard to produce specifications in many application domains (e.g., planning, business process management, run-time monitoring, reactive synthesis). Several studies approached the respective satisfiability problem. In this paper, we investigate the problem of extracting the unsatisfiable core in LTLf specifications. We provide four algorithms for extracting an unsatisfiable core leveraging the adaptation of state-of-the-art approaches to LTLf satisfiability checking. We implement the different approaches within the respective tools and carry out an experimental evaluation on a set of reference benchmarks, restricting to the unsatisfiable ones. The results show the feasibility, effectiveness, and complementarities of the different algorithms and tools.

Abstract--Abstract Visual Reasoning (AVR) problems are commonly used to approximate human intelligence. They test the ability of applying previously gained knowledge, experience and skills in a completely new setting, which makes them particularly well-suited for this task. Recently, the AVR problems have become popular as a proxy to study machine intelligence, which has led to emergence of new distinct types of problems and multiple benchmark sets. In this work we review this emerging AVR research and propose a taxonomy to categorise the AVR tasks along 5 dimensions: input shapes, hidden rules, target task, cognitive function, and specific challenge. The perspective taken in this survey allows to characterise AVR problems with respect to their shared and distinct properties, provides a unified view on the existing approaches to solving AVR tasks, shows how the AVR problems relate to practical applications, and outlines promising directions for future work. One of them refers to the observation that in the machine learning literature different tasks are considered in isolation, which is in the stark contrast with the way the AVR tasks are used to measure human intelligence, where multiple types of problems are combined within a single IQ test.