Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (3 results)

  • Thesis
  • Peer Reviewed
UC San Diego Electronic Theses and Dissertations (2018)

The field of optimal transportation is a broad area of theory pertaining to the computation of a mapping to transform one probability distribution into another. Many computational strategies solving the transport problem spanning decades of research have led to several interesting applications in areas such as statistics, economics, machine learning and computer science. However, there also exist many limitations to these algorithms, including the ever-present difficulty of scaling to larger datasets, both with respect to dimension and number of available samples from which we would like to extract useful information. Furthermore, although we have seen a dramatic increase in the availability of powerful distributed computational resources throughout the past few decades, such as publicly available CPU clusters and GPU compute nodes, many modern algorithmic approaches to the transport problem are not specifically designed with parallelism in mind to accommodate such frameworks. In a world where the continuous interaction between human users and distributed computational resources drives our technologically modern lives, this capability is critical, especially with applications that are expected to work in real-time.

Building upon previous research from our group, this work investigates a parallelized, computational problem to create optimal transport maps that is designed with the notion of remote scalability in mind; this work focuses on a CUDA-based implementation of the framework that utilizes GPU computational resources to accommodate analysis of data in higher dimensions not often seen in the field of computational optimal transportation, and that furthermore scale with the availability of additional hardware. We will also present applications using this framework in several different areas of machine learning with an emphasis on human-computer interfaces, including a novel multi-user brain-computer interface, a classification problem pertaining to automated sleep-staging based on electroencephalographical (EEG) data, and an example of generative modeling using the MNIST dataset. Finally, we'll discuss a future direction for application of this framework to preference-based deep reinforcement learning.

    Cover page: Scalable Measure Transportation and Applications in Machine Learning and Human Computer Interfaces
    • Article
    • Peer Reviewed
    UC San Diego Previously Published Works (2021)
    Machine learning has the potential to change the practice of medicine, particularly in areas that require pattern recognition (e.g. radiology). Although automated classification is unlikely to be perfect, few modern machine learning tools have the ability to assess their own classification confidence to recognize uncertainty that might need human review. Using automated single-channel sleep staging as a first implementation, we demonstrated that uncertainty information (as quantified using Shannon entropy) can be utilized in a "human in the loop" methodology to promote targeted review of uncertain sleep stage classifications on an epoch-by-epoch basis. Across 20 sleep studies, this feedback methodology proved capable of improving scoring agreement with the gold standard over automated scoring alone (average improvement in Cohen's Kappa of 0.28), in a fraction of the scoring time compared to full manual review (60% reduction). In summary, our uncertainty-based clinician-in-the-loop framework promotes the improvement of medical classification accuracy/confidence in a cost-effective and economically resourceful manner.
      Cover page: Statistical uncertainty quantification to augment clinical decision support: a first implementation in sleep medicine
      • Article
      • Peer Reviewed
      UC San Diego Previously Published Works (2020)
      Glaucoma, the leading cause of irreversible blindness, affects >70 million people worldwide. Lowering intraocular pressure via topical administration of eye drops is the most common first-line therapy for glaucoma. This treatment paradigm has notoriously high non-adherence rates: ranging from 30% to 80%. The advent of smart phone enabled technologies creates promise for improving eyedrop adherence. However, previous eyedrop electronic monitoring solutions had awkward medication bottle adjuncts and crude software for monitoring the administration of a drop that adversely affected their ability to foster sustainable improvements in adherence. The current work begins to address this unmet need for wireless technology by creating a "smart drop" bottle. This medication bottle is instrumented with sensing electronics that enable detection of each eyedrop administered while maintaining the shape and size of the bottle. This is achieved by a thin electronic force sensor wrapped around the bottle and underneath the label, interfaced with a thin electronic circuit underneath the bottle that allows for detection and wireless transmission to a smart-phone application. We demonstrate 100% success rate of wireless communication over 75 feet with <1% false positive and false negative rates of single drop deliveries, thus providing a viable solution for eyedrop monitoring for glaucoma patients.
        Cover page: Smart Electronic Eyedrop Bottle for Unobtrusive Monitoring of Glaucoma Medication Adherence
        Top