Category: Uncategorized

  • Supervised Hierarchical Clustering Using Transformed LSTM Networks

    Supervised Hierarchical Clustering Using Transformed LSTM Networks – We train a recurrent neural network to learn the relation between two images and combine them in a new image-to-image matching task. To learn the relation between images and images, we used a simple, yet powerful feature-based representation. In our experiments, we use an extensive dataset to […]

  • Learning the Structure and Parameters of Structured Classifiers and Prostate Function Prediction Models

    Learning the Structure and Parameters of Structured Classifiers and Prostate Function Prediction Models – We present the method of using the concept of a causal model to perform probabilistic inference under a supervised learning paradigm. The method is based on constructing a model that is invariant to an unknown data set and using the model […]

  • Loss Functions for Robust Gaussian Processes with Noisy Path Information

    Loss Functions for Robust Gaussian Processes with Noisy Path Information – In this paper, we propose a new method to learn a probabilistic model of a probabilistic data base from a probabilistic model of the data, called Gaussian Processes with Noisy Path Information (GP-PEPHiP). GP-PEPH is a probabilistic model of an unimportant data base where […]

  • Adaptive Dynamic Mode Decomposition of Multispectral Images for Depth Compensation in Unstructured Sensor Data

    Adaptive Dynamic Mode Decomposition of Multispectral Images for Depth Compensation in Unstructured Sensor Data – Deep learning is a machine learning technique that makes use of deep neural networks (DNNs). In this paper, we describe how the deep network architecture can be used for a class of image classification tasks, including the classification of images. […]

  • Learning Image Representation for Complex Problems

    Learning Image Representation for Complex Problems – We propose a novel deep learning framework for solving a multi-instance visual system task, called multi-instance object segmentation (MID). The MID framework is a joint convolutional recurrent network (CNN), a convolutional hierarchical recurrent network (CNN) and an input-to-output network (IRNN). By combining the proposed CNN and IRNN, the […]

  • Proceedings of the 38th Annual Workshop of the Austrian Machine Learning Association (ÖDAI), 2013

    Proceedings of the 38th Annual Workshop of the Austrian Machine Learning Association (ÖDAI), 2013 – This paper presents a new tool to analyze and evaluate the performance of the state-of-the-art deep neural networks (DNNs). In fact, the traditional method of the state-of-the-art DNNs is to design a model on the data manifold without analyzing the […]

  • Constrained Multi-View Image Classification with Multi-temporal Deep CNN Regressions

    Constrained Multi-View Image Classification with Multi-temporal Deep CNN Regressions – We propose a novel non-Gaussianized Multi-views (MMTV) approach for high-dimensional object localization. An important limitation in MMTV, for two reasons, is that the non-Gaussianity of the features is not well studied in the non-Gaussianity data. To address this problem we propose a novel multi-view data […]

  • Using the Multi-dimensional Bilateral Distribution for Textual Discrimination

    Using the Multi-dimensional Bilateral Distribution for Textual Discrimination – We present a new dataset for a novel kind of semantic discrimination (tense) task aiming at comparing two types of text: semantic and unsemantically. It includes large-scale annotated hand annotated datasets that are large in size and are capable of covering an entire language. We propose […]

  • Semantic Machine Meet Benchmark

    Semantic Machine Meet Benchmark – In this work we present a new deep learning technique for semantic object detection and tracking in an image-based 3D scene system. The proposed approach relies on a hierarchical deep neural network (DNN). The hierarchical DNN models the scene by selecting the scenes and identifying the relevant object categories according […]

  • Learning Language Models from Texts via Iterative Estimation

    Learning Language Models from Texts via Iterative Estimation – We present a fully scalable algorithm to extract a new representation of semantic word embeddings for the Chinese Wikipedia text. Our method achieves state-of-the-art performance on benchmark datasets for English and Chinese-English semantic word embeddings. Our system runs in real time for approximately half an hour […]