Machine learning methods have a groundbreaking impact in many application domains, but their application on real robotic platforms is still limited. Despite the many challenges associated with combining machine learning technology with robotics, robot learning remains one of the most promising directions for enhancing thecapabilities ofrobots.

Tuning hyperparameters is a crucial but arduous part of the machine learning pipeline. Hyperparameter optimization is even more challenging in federated learning, where models are learned over a distributed network of heterogeneous devices; here, the need to keep data on device and perform local training makes it difficult to efficiently train and evaluate configurations. In this work, we investigate the problem of federated hyperparameter tuning. We first identify key challenges and show how standard approaches may be adapted to form baselines for the federated setting. Then, by making a novel connection to the neural architecture search technique of weight-sharing, we introduce a new method, FedEx, to accelerate federated hyperparameter tuning that is applicable to widely-used federated optimization methods such as FedAvg and recent variants. Theoretically, we show that a FedEx variant correctly tunes the on-device learning rate in the setting of online convex optimization across devices. Empirically, we show that FedEx can outperform natural baselines for federated hyperparameter tuning by several percentage points on the Shakespeare, FEMNIST, and CIFAR-10 benchmarks--obtaining higher accuracy using the same training budget.

Membership in ACM includes a subscription to Communications of the ACM (CACM), the computing industry's most trusted source for staying connected to the world of advanced computing. The Regional Special Section of the Arab World highlights some of the region's exciting, innovative, and socially relevant advances in computing and its applications. It is with great pleasure that we present this Communications of the ACM Regional Special Section of the Arab World. In this second edition, we highlight some of the region's exciting, innovative, and socially relevant advances in computing and its applications. The Arab world is home to a rich mosaic of cultures, histories, and geographies, stretching from the Atlantic Ocean to the Gulf.

How can we attribute the behaviors of machine learning models to their training data? While the classic influence function sheds light on the impact of individual samples, it often fails to capture the more complex and pronounced collective influence of a set of samples. To tackle this challenge, we study the Most Influential Subset Selection (MISS) problem, which aims to identify a subset of training samples with the greatest collective influence. We conduct a comprehensive analysis of the prevailing approaches in MISS, elucidating their strengths and weaknesses. Our findings reveal that influence-based greedy heuristics, a dominant class of algorithms in MISS, can provably fail even in linear regression.

Despite extensive efforts to create fairer machine learning (ML) datasets, there remains a limited understanding of the practical aspects of dataset curation. Drawing from interviews with 30 ML dataset curators, we present a comprehensive taxonomy of the challenges and trade-offs encountered throughout the dataset curation lifecycle. Our findings underscore overarching issues within the broader fairness landscape that impact data curation. We conclude with recommendations aimed at fostering systemic changes to better facilitate fair dataset curation practices.

Past analyses of reinforcement learning from human feedback (RLHF) assume that the human evaluators fully observe the environment. What happens when human feedback is based only on partial observations? We formally define two failure cases: deceptive inflation and overjustification. Modeling the human as Boltzmann-rational w.r.t. a belief over trajectories, we prove conditions under which RLHF is guaranteed to result in policies that deceptively inflate their performance, overjustify their behavior to make an impression, or both. Under the new assumption that the human's partial observability is known and accounted for, we then analyze how much information the feedback process provides about the return function.

The past few years have seen an explosion in the use of artificial intelligence on embedded and edge devices. Starting with the keyword spotting models that wake up the digital assistants built into every modern cellphone, "edge AI" products have made major inroads into our homes, wearable devices, and industrial settings. They represent the application of machine learning to a new computational context. ML practitioners are the champions at building datasets, experimenting with different model architectures, and building best-in-class models. ML experts also understand the potential of machine learning to transform the way that humans and technology work together.

Are you ready to take your business to the next level? Look no further than the rapidly expanding field of artificial intelligence. With advancements in technology, AI is revolutionizing industries and creating new opportunities for entrepreneurs. From healthcare to finance, AI is improving efficiency and driving innovation. Entrepreneurs who embrace this technology can create new business models and revenue streams.

Since neural networks learn by making mistakes, they require enormous volumes of training data. It's no accident that neural networks only gained popularity after most businesses adopted big data analytics and gathered enormous data repositories. The data used during the training stage must be labeled so the model can determine if its informed estimate was correct because the model's initial iterations entail making educated guesses about the contents of an image or sections of speech. This indicates that even though many businesses using big data have a lot of data, unstructured data is less useful. Deep learning models cannot be taught on unstructured data, hence unstructured data can only be examined by a deep learning model once it has been trained and achieves an acceptable degree of accuracy.

In cooperative multi-agent reinforcement learning (MARL), due to its on-policy nature, policy gradient (PG) methods are typically believed to be less sample efficient than value decomposition (VD) methods, which are off-policy. However, some recent empirical studies demonstrate that with proper input representation and hyper-parameter tuning, multi-agent PG can achieve surprisingly strong performance compared to off-policy VD methods. Why could PG methods work so well? In this post, we will present concrete analysis to show that in certain scenarios, e.g., environments with a highly multi-modal reward landscape, VD can be problematic and lead to undesired outcomes. In addition, PG methods with auto-regressive (AR) policies can learn multi-modal policies.