Tag: deep learning

Abstract: We present a histogram layer for artificial neural networks (ANNs). An essential aspect of texture analysis is the extraction of features that describe the distribution of values in local spatial regions. The proposed histogram layer directly computes the spatial distribution of features for texture analysis and parameters for the layer are estimated during backpropagation. […]

Congratulations to our labmate, Sarah Walker! Sarah presented her work, titled “Divergence Regulated Encoder Network for Joint Dimensionality Reduction and Classification” at UF’s 2021 Undergraduate Research Virtual Symposium.  The virtual symposium featured outstanding undergraduate researchers across all colleges at UF. You can get more information about Sarah’s talk here and can check out the paper […]

Abstract: In this paper, we investigate performing joint dimensionality reduction and classification using a novel histogram neural network. Motivated by a popular dimensionality reduction approach, t-Distributed Stochastic Neighbor Embedding (t-SNE), our proposed method incorporates a classification loss computed on samples in a low-dimensional embedding space. We compare the learned sample embeddings against coordinates found by […]

Congratulations to our labmates: Sarah Walker, Joshua Peeples, Jeff Dale, James Keller and Alina Zare!  Their paper, “Explainable Systematic Analysis for Synthetic Aperture Sonar Imagery” was recently accepted to the IEEE International Geoscience and Remote Sensing Symposium (IGARSS). In their paper, the authors provide an in-depth analysis to the factors that affect performance of texture […]

Abstract: In this work, we present an in-depth and systematic analysis using tools such as local interpretable model-agnostic explanations (LIME) and divergence measures to analyze what changes lead to improvement in performance in fine tuned models for synthetic aperture sonar (SAS) data. We examine the sensitivity to factors in the fine tuning process such as […]

Abstract: Forests provide biodiversity, ecosystem, and economic services. Information on individual trees is important for understanding forest ecosystems but obtaining individual-level data at broad scales is challenging due to the costs and logistics of data collection. While advances in remote sensing techniques allow surveys of individual trees at unprecedented extents, there remain technical challenges in […]

Congratulations to our labmates, Matt Cook, Alina Zare and Paul Gader for presenting at the 37th International Conference on Machine Learning (ICML) Workshop on Uncertainty and Robustness in Machine Learning! Their paper, titled “Outlier Detection through Null Space Analysis of Neural Networks”, introduces a novel method for detecting outliers in a set of data. Matt will […]

  We are happy to announce the publication of a new dataset!  The NEON Tree Crowns Dataset is a collection of individual tree crown estimates for 100 million trees from 37 geographic sites across the United States.  This dataset provides predicted bounding boxes, tree heights, crown areas, class labels and confidence scores for images taken from […]

Abstract: The NeonTreeCrowns dataset is a set of individual level crown estimates for 100 million trees at 37 geographic sites across the United States surveyed by the National Ecological Observation Network’s Airborne Observation Platform. Each rectangular bounding box crown prediction includes height, crown area, and spatial location. Links: Citation: B. Weinstein, S. Marconi, A. Zare, […]

Abstract: Many machine learning classification systems lack competency awareness. Specifically, many systems lack the ability to identify when outliers (e.g., samples that are distinct from and not represented in the training data distribution) are being presented to the system. The ability to detect outliers is of practical significance since it can help the system behave […]