Active learning enables efficient model training by leveraging interactions between machine learning agents and human annotators. We study and propose a novel framework that formulates batch active learning from the sparse approximation's perspective. Our active learning method aims to find an informative subset from the unlabeled data pool such that the corresponding training loss function approximates its full data pool counterpart. We realize the framework as sparsity-constrained discontinuous optimization problems, which explicitly balance uncertainty and representation for large-scale applications and could be solved by greedy or proximal iterative hard thresholding algorithms. The proposed method can adapt to various settings, including both Bayesian and non-Bayesian neural networks. Numerical experiments show that our work achieves competitive performance across different settings with lower computational complexity.

Learning curves provide insight into the dependence of a learner's generalization performance on the training set size. This important tool can be used for model selection, to predict the effect of more training data, and to reduce the computational complexity of model training and hyperparameter tuning. This review recounts the origins of the term, provides a formal definition of the learning curve, and briefly covers basics such as its estimation. Our main contribution is a comprehensive overview of the literature regarding the shape of learning curves. We discuss empirical and theoretical evidence that supports well-behaved curves that often have the shape of a power law or an exponential. We consider the learning curves of Gaussian processes, the complex shapes they can display, and the factors influencing them. We draw specific attention to examples of learning curves that are ill-behaved, showing worse learning performance with more training data. To wrap up, we point out various open problems that warrant deeper empirical and theoretical investigation. All in all, our review underscores that learning curves are surprisingly diverse and no universal model can be identified.

We present a new NLP task and dataset from the domain of the U.S. civil procedure. Each instance of the dataset consists of a general introduction to the case, a particular question, and a possible solution argument, accompanied by a detailed analysis of why the argument applies in that case. Since the dataset is based on a book aimed at law students, we believe that it represents a truly complex task for benchmarking modern legal language models. Our baseline evaluation shows that fine-tuning a legal transformer provides some advantage over random baseline models, but our analysis reveals that the actual ability to infer legal arguments remains a challenging open research question.

In this paper, we increase the availability and integration of devices in the learning process to enhance the convergence of federated learning (FL) models. To address the issue of having all the data in one location, federated learning, which maintains the ability to learn over decentralized data sets, combines privacy and technology. Until the model converges, the server combines the updated weights obtained from each dataset over a number of rounds. The majority of the literature suggested client selection techniques to accelerate convergence and boost accuracy. However, none of the existing proposals have focused on the flexibility to deploy and select clients as needed, wherever and whenever that may be. Due to the extremely dynamic surroundings, some devices are actually not available to serve as clients in FL, which affects the availability of data for learning and the applicability of the existing solution for client selection. In this paper, we address the aforementioned limitations by introducing an On-Demand-FL, a client deployment approach for FL, offering more volume and heterogeneity of data in the learning process. We make use of the containerization technology such as Docker to build efficient environments using IoT and mobile devices serving as volunteers. Furthermore, Kubernetes is used for orchestration. The Genetic algorithm (GA) is used to solve the multi-objective optimization problem due to its evolutionary strategy. The performed experiments using the Mobile Data Challenge (MDC) dataset and the Localfed framework illustrate the relevance of the proposed approach and the efficiency of the on-the-fly deployment of clients whenever and wherever needed with less discarded rounds and more available data.

Flexible task planning is still a significant challenge for robots. The inability of robots to creatively adapt their task plans to new or unforeseen challenges is largely attributable to their limited understanding of their activities and the environment. Cooking, for example, requires a person to occasionally take risks that a robot would find extremely dangerous. We may obtain manipulation sequences by employing knowledge that is drawn from numerous video sources thanks to knowledge retrieval through graph search.

Today, no company can survive in the market without using Machine Learning models, and clients will not purchase from companies that do not offer ML-enhanced services. Making a Machine Learning ecosystem operational can help turn enterprise data into a predictive engine for the company. Data and analytics leaders have always understood the benefits of using Machine Learning (ML) for their businesses. The value mostly comes in three ways: operational efficiencies, better top-line growth, and enhanced employee and customer experiences. To unlock that value, however, line-of-business teams must overcome several persistent challenges, with the biggest one being their inability to draw insights from the vast quantities of data they possess.

AI is behind many small advances such as optimising the ripening of fruit, an artificial sense of smell, a system that improves student performance, learning more about Don Quixote or creating a doll to accompany our elders. These are curious and lesser-known examples of how artificial intelligence can be applied to our benefit on a smaller scale. Fifteen examples of this are worth mentioning. The Tekniker technology centre, located in the Basque Country, has developed a tool to detect characteristics or defects in bananas and optimise their ripening through image processing based on artificial intelligence. Image processing based on neural networks is an artificial intelligence technique that can help develop solutions for sectors such as the food industry to identify errors in production and processing and improve product quality.

"Tell me and I forget, teach me and I may remember, involve me and I learn." The pandemic not only propelled us to rethink the means of how we learn – accelerating the adoption of online learning, but also brought to the forefront discourses around the quality of education, and more importantly, accessibility for all. Adoption of emerging technologies such as Artificial Intelligence, Augmented Reality and Virtual Reality have enabled us to provide students with immersive and personalised learning experiences and, while doing so, achieve superior learning outcomes. And to my mind, adoption of AI bots as virtual teachers is the next big innovation that has the potential to transform the face of education. AI bots, chatbots in particular, have become a rather common phenomenon in today's time.

Music classroom with AI generated MIDI. MIDI makers on Youtube have millions of followers for music lessons. However, a musician must recognize melodies, chords, and emotions from the original audio and convert them into musical performances. Therefore, creating a MIDI is a creative endeavor that requires musical understanding. Today you can generate a MIDI from your favorite song without any help.

The University of Florida, a proponent for ethics in artificial intelligence, is part of a new global agreement with seven other worldwide universities that are committed to the development of human-centered approaches to artificial intelligence (AI) that will impact people everywhere. During the Global University Summit at Notre Dame University, Joseph Glover, UF provost and senior vice president of academic affairs, signed The Rome Call for AI Ethics on October 27 on behalf of the University of Florida and served as a panelist for the two-day summit attended by 36 universities invited from around the world. The event was held in Notre Dame, IN. The signing indicates a commitment to the principles of the Rome Call for AI Ethics: to ensure artificial intelligence serves the interests of humanity and to support regulations and principles to deliver emerging technologies that are ethically centered. UF joins a network of universities that will share best practices, tools, and educational content, as well as meet regularly to share updates and discuss innovative ideas.