This article surveys the use of natural language in robotics from a robotics point of view. To use human language, robots must map words to aspects of the physical world, mediated by the robot's sensors and a...
This article surveys the use of natural language in robotics from a robotics point of view. To use human language, robots must map words to aspects of the physical world, mediated by the robot's sensors and actuators. This problem differs from other natural language processing domains due to the need to ground the language to noisy percepts and physical actions. Here, we describe central aspects of language use by robots, including understanding natural language requests, using language to drive learning about the physical world, and engaging in collaborative dialogue with a human partner. We describe common approaches, roughly divided into learning methods, logic-based methods, and methods that focus on questions of human–robot interaction. Finally, we describe several application domains for language-using robots.
New technologies have led to vast troves of large and complex data sets across many scientific domains and industries. People routinely use machine learning techniques not only to process, visualize, and make predicti...
New technologies have led to vast troves of large and complex data sets across many scientific domains and industries. People routinely use machine learning techniques not only to process, visualize, and make predictions from these big data, but also to make data-driven discoveries. These discoveries are often made using interpretable machine learning, or machine learning models and techniques that yield human-understandable insights. In this article, we discuss and review the field of interpretable machine learning, focusing especially on the techniques, as they are often employed to generate new knowledge or make discoveries from large data sets. We outline the types of discoveries that can be made using interpretable machine learning in both supervised and unsupervised settings. Additionally, we focus on the grand challenge of how to validate these discoveries in a data-driven manner, which promotes trust in machine learning systems and reproducibility in science. We discuss validation both from a practical perspective, reviewing approaches based on data-splitting and stability, as well as from a theoretical perspective, reviewing statistical results on model selection consistency and uncertainty quantification via statistical inference. Finally, we conclude byhighlighting open challenges in using interpretable machine learning techniques to make discoveries, including gaps between theory and practice for validating data-driven discoveries.
Coordination between different sensory systems is a necessary element of sensory processing. Where and how signals from different sense organs converge onto common neural circuitry have become topics of increasing int...
Coordination between different sensory systems is a necessary element of sensory processing. Where and how signals from different sense organs converge onto common neural circuitry have become topics of increasing interest in recent years. In this article, we focus specifically on visual–auditory interactions in areas of the mammalian brain that are commonly considered to be auditory in function. The auditory cortex and inferior colliculus are two key points of entry where visual signals reach the auditory pathway, and both contain visual- and/or eye movement–related signals in humans and other animals. The visual signals observed in these auditory structures reflect a mixture of visual modulation of auditory-evoked activity and visually driven responses that are selective for stimulus location or features. These key response attributes also appear in the classic visual pathway but may play a different role in the auditory pathway: to modify auditory rather than visual perception. Finally, while this review focuses on two particular areas of the auditory pathway where this question has been studied, robust descending as well as ascending connections within this pathway suggest that undiscovered visual signals may be present at other stages as well.
The Human Genome Project was an enormous accomplishment, providing a foundation for countless explorations into the genetics and genomics of the human species. Yet for many years, the human genome reference sequence r...
The Human Genome Project was an enormous accomplishment, providing a foundation for countless explorations into the genetics and genomics of the human species. Yet for many years, the human genome reference sequence remained incomplete and lacked representation of human genetic diversity. Recently, two major advances have emerged to address these shortcomings: complete gap-free human genome sequences, such as the one developed by the Telomere-to-Telomere Consortium, and high-quality pangenomes, such as the one developed by the Human Pangenome Reference Consortium. Facilitated by advances in long-read DNA sequencing and genome assembly algorithms, complete human genome sequences resolve regions that have been historically difficult to sequence, including centromeres, telomeres, and segmental duplications. In parallel, pangenomes capture the extensive genetic diversity across populations worldwide. Together, these advances usher in a new era of genomics research, enhancing the accuracy of genomic analysis, paving the path for precision medicine, and contributing to deeper insights into human biology.
暂无评论