Search Constraints

Start Over
Filtering by: Creator Hummel, Andrew J. Remove constraint Creator: Hummel, Andrew J. Committee Member Fern, Alan Remove constraint Committee Member: Fern, Alan

Search Results

  • Methods for Detection and Recovery of Out-of-Distribution Examples

    Creator:
    Ratzlaff, Neale J.
    Abstract:
    Deep neural networks currently comprise the backbone of many applications where safety is a critical concern, for example: autonomous driving and medical diagnostics. Unfortunately these systems currently fail to detect out-of-distribution (OOD) inputs and can be prone to making dangerous errors when exposed to them. In addition, these same systems...
    Resource Type:
    Masters Thesis
    Full Text:
    Ij is a j × j identity matrix, and generate parameters using a generator from s. Figure 2.1 shows

  • Incorporating labeled features into image classification using generalized expectation

    Creator:
    Moore, Travis
    Abstract:
    Image classification is a difficult problem, often requiring large training sets to get satisfactory results. However this is a task that humans perform very well, and incorporating user feedback into these learning algorithms could help reduce the dependency on large amounts of labeled training data. This process has already been...
    Resource Type:
    Masters Thesis
    Full Text:
    , if a segment has a SIFT vector s, and it is clustered into group j, then we say that the feature

  • Open Set Learning with Counterfactual Images

    Creator:
    Neal, Lawrence W.
    Abstract:
    In open set recognition, a classifier must label instances of known classes while detecting instances of unknown classes not encountered during training. To detect unknown classes while still generalizing to new instances of existing classes, this thesis introduces a dataset augmentation technique called counterfactual image generation. This approach, based on...
    Resource Type:
    Masters Thesis
    Full Text:
    examples. Given a mean activation vector for each class j ∈ [1...K], a Weibull distribution with values (τ

  • Efficient training and feature induction in sequential supervised learning

    Creator:
    Hao, Guohua
    Abstract:
    Sequential supervised learning problems arise in many real applications. This dissertation focuses on two important research directions in sequential supervised learning: efficient training and feature induction. In the direction of efficient training, we study the training of conditional random fields (CRFs), which provide a flexible and powerful model for sequential...
    Resource Type:
    Dissertation
    Full Text:
    as P (xt|yt) = ∏ j P (xt,j|yt) . This assumption of independence of each input feature xt,j

  • Fine-Grained Object Recognition Under Limited Training Data

    Creator:
    Lam, Michael Q.
    Abstract:
    This dissertation addresses object recognition in challenging settings, where distinct object classes are visually very similar (e.g., species of birds and insects) and/or access to training examples of object classes is limited (e.g., due to the associated high costs of data annotation). In this dissertation, we present a variety of...
    Resource Type:
    Dissertation
    Full Text:
    subsequent iterations j, we perform search using a mixture of the learned heuristic from the previous

  • Anomaly Detection and Probabilistic Diagnosis for Automated Data Quality Control

    Creator:
    Zemicheal, Tadesse H.
    Abstract:
    Advances in sensor technology are greatly expanding the range of quantities that can be measured while simultaneously reducing the cost. However, deployed sensors drift out of calibration and fail, so every sensor network requires quality control procedures to promptly detect these failures. To address these problems, we propose a two-level...
    Resource Type:
    Dissertation
    Full Text:
    . Among them, I would like to give my especial thank Amran Siddique, Si Liu, Michael Slater, Andrew

  • Bayesian optimization with empirical constraints

    Creator:
    Azimi, Javad
    Abstract:
    Bayesian Optimization (BO) methods are often used to optimize an unknown function f(•) that is costly to evaluate. They typically work in an iterative manner. In each iteration, given a set of observation points, BO algorithms select k ≥ 1 points to be evaluated. The results of those points are...
    Resource Type:
    Dissertation
    Full Text:
    captured by: J(Q) = E(x1,···,x|Q|) [ E(y1,···,y|Q|) [ max { y1, . . . , y|Q| } ∣∣∣D, (x1, · · ·, x|Q

  • Rare category detection using hierarchical mean shift

    Creator:
    Vatturi, Pavan Kumar
    Abstract:
    Many applications in surveillance, monitoring, scientific discovery, and data cleaning require the identification of anomalies. Although many methods have been developed to identify statistically significant anomalies, a more difficult task is to identify anomalies that are both interesting and statistically significant. Category detection is an emerging area of machine learning...
    Resource Type:
    Masters Thesis
    Full Text:
    ][n]) = SemiSupervisedEM (S, Labels);3 where PointOwnership[i][j] is an estimate of the degree of

  • Efficient Algorithms for Robust Spatiotemporal Data Analysis

    Creator:
    Moore, Travis W.
    Abstract:
    Many large-scale data analysis applications involve data that can vary over both time and space. Often the primary goal of analyzing spatiotemporal data is identifying trends, movements, and sudden changes with respect to time, location, or both. This can include a variety of applications in economics (housing prices, unemployment, job...
    Resource Type:
    Dissertation
    Full Text:
    a row selector matrix, where K[j,j] = 1 if point j is in C, and all other values are zero. For a

  • Uncertainty in Deep Learning with Implicit Neural Networks

    Creator:
    Ratzlaff, Neale
    Abstract:
    The ability to extract uncertainties from predictions is crucial for the adoption of deep learning systems to safety-critical applications. Uncertainty estimates can be used as a failure signal, which is necessary for automating complex tasks where safety is a concern. Furthermore, current deep learning systems do not provide uncertainty estimates,...
    Resource Type:
    Dissertation
    Full Text:
    covariance, and input to layer-wise generators {f1, · · · , fL}. Each generator fj outputs parameters θj for