Search for: All records

In this opinion piece, the authors, from the fields of artificial intelligence (AI) and psychology, reflect on how the foundational discoveries of Nobel laureates Hopfield and Hinton have influenced their research. They also discuss emerging directions in AI and the challenges that lie ahead for neural networks and machine learning. 

Visual learning often occurs in a specific context, where an agent acquires skills through exploration and tracking of its location in a consistent environment. The historical spatial context of the agent provides a similarity signal for self-supervised contrastive learning. We present a unique approach, termed Environmental Spatial Similarity (ESS), that complements existing contrastive learning methods. Using images from simulated, photorealistic environments as an experimental setting, we demonstrate that ESS outperforms traditional instance discrimination approaches. Moreover, sampling additional data from the same environment substantially improves accuracy and provides new augmentations. ESS allows remarkable proficiency in room classification and spatial prediction tasks, especially in unfamiliar environments. This learning paradigm has the potential to enable rapid visual learning in agents operating in new environments with unique visual characteristics. Potentially transformative applications span from robotics to space exploration. Our proof of concept demonstrates improved efficiency over methods that rely on extensive, disconnected datasets. 

Data science, an emerging multidisciplinary field, resides at the intersec- tion of computational sciences, statistical modeling, and domain-specific sciences. The current norm for data science education predominantly focuses on graduate programs, which presume a pre-existing knowledge base in one or more relevant sciences. However, this framework often overlooks those who don’t plan to pursue graduate studies, thereby limiting their exposure to this rapidly expanding field. Penn State addressed this gap by establishing one of the first undergraduate degree programs in Data Sciences, a collaboration between the College of Information Sci- ences and Technology, the Department of Computer Science and Engineering, and the Department of Statistics. One key component of this program is a project-focused, writing-intensive course designed for upper-class undergraduates. This course guides students through the entire data science project pipeline, from problem identifica- tion to solution presentation. It allows students to apply foundational data science principles to real-world problems, advancing their understanding through practi- cal application. This chapter details the objectives, rationale, and course design, alongside reflections from our teaching experience. The insights provided could be helpful to instructors developing similar data science programs or courses at an undergraduate level, broadening the influence of this important field