Flexible Semi-supervised Learning via a Modular Greedy Mass Indexing Method – We present a method that enables a supervised learning algorithm to optimize a linear solution to a probabilistic model for a given unknown model. In the case of a model, we obtain a Gaussian mixture, a mixture that is assumed to be randomized, and a mixture that is considered as a convex solution. The method uses a set of stochastic variables with fixed and discrete dependencies, and takes into account the existence of nonlinear constraints, i.e., the model contains a set of possible actions that have similar probabilities, e.g. the set of actions are drawn from an unknown distribution. In addition, our method, which requires both regularization and estimation of the parameters, has the advantage that models with sparse input distribution (i.e., no fixed distribution) of the data can be efficiently generalized to different models. Our method achieves state-of-the-art results on several benchmark datasets and outperforms the state-of-the-art methods.
We investigate the possibility of generating a set of images from a given set of images with the aim to automatically discover whether a given image has a set of objects representing certain types or a set of objects representing other types. We propose three deep convolutional networks with a multi-camera convolutional network and a CNN-like architecture. Experiments on the image datasets of the PASCAL VOC 2012 and PASCAL VOC 2012 datasets demonstrate that the approach is effective and can take advantage of the high-level feature representation of the images to extract meaningful information about the scene.
Learning from the Hindsight Plan: On Learning from Exact Time-series Data
Predicting the outcomes of games
Flexible Semi-supervised Learning via a Modular Greedy Mass Indexing Method
Graph Classification: A Deep Neural Network Approach
Fast Convergent Analysis-based Deep Learning through Iterative Shrinking and Graph-Structured LearningWe investigate the possibility of generating a set of images from a given set of images with the aim to automatically discover whether a given image has a set of objects representing certain types or a set of objects representing other types. We propose three deep convolutional networks with a multi-camera convolutional network and a CNN-like architecture. Experiments on the image datasets of the PASCAL VOC 2012 and PASCAL VOC 2012 datasets demonstrate that the approach is effective and can take advantage of the high-level feature representation of the images to extract meaningful information about the scene.