A variety of important machine learning applications require predictions on test data with different characteristics than the data on which a model was trained and validated. In particular, test data may have a different relative frequency of positives and negatives (i.e., class distribution) and/or different mislabeling costs of false positive...
This paper presents an overview of cogeneration, also known as combined heat and power (CHP). The technology of cogeneration is described and the three primary cogeneration processes are discussed. The transition between parallel operation with the utility and stand-alone operation is discussed. The tradeoff between a synchronous and an induction...
Using supervised machine learning (ML) to train a computer vision model typically requires human annotators to label objects in images and video. Given a large training dataset, this can be labor intensive, presenting a significant bottleneck in the model-development process. LabelFlicks is an open-source desktop application that aims to address...
Personalization is defined as a process that facilitates interaction among consumers and providers such that individual consumers are enabled to more readily access the content and services of providers, and individual providers are enabled to more effectively and easily deliver their content and services to consumers. This project presents a...
Question Answering in natural language processing has achieved significant progress in recent years. Yet, training and testing set methodology to evaluate the language models has proved inadequate. Adversarial examples aid us in finding loopholes inside these models and provide insights into their inner workings. In this work, an evaluation based...
This paper addresses the high model complexity and overconfident frame labeling of state-of-the-art (SOTA) action segmenters. Their complexity is typically justified by the need to sequentially refine action segmentation through multiple stages of a deep architecture. However, this multistage refinement does not take into account uncertainty of frame labeling predicted...
The OSU Motor Systems Resource Facility (MSRF), co-directed by the Principal Investigators (PI's, bios included), is an Energy Systems Laboratory with operating capabilities up to 750kV A, testbeds up to 300hp, a 120kVA fully programmable source, and a bi-directional grid interface enabling regeneration back onto the grid. The MSRF was...