Learning Discrete Dynamical Systems Based on Interacting with the Information Displacement Model – Many deep learning methods have been proposed and evaluated on a few domains. In this paper, we propose Deep Neural Network (DNN) models for the object recognition task. We first show that, in most cases, deep networks can achieve accuracies comparable to neural networks, but have a much larger computational cost. We suggest that deep DNN models are at least as computationally efficient as state-of-the-art deep networks. Our model is based on Deep Convolutional Neural Network (DCNN). We give the best experimental performance on the standard datasets (MALE (MCA-12), MEDIA (MCA-8), and COCO (COCO-8), using a large amount of data. We also give a theoretical analysis to show that the use of deep DCNN is a good policy. The proposed models are evaluated against the state-of-the-art models for object recognition and classify the results for these two tasks. The proposed DNN models can be applied to different domain.
We present a nonlinear model to model the temporal evolution of human knowledge about the world. Our approach is to first embed temporally related knowledge into the form of a multidimensional variable. We then embed the inter- and intra-variable covariate into a multidimensional structure in order to model the temporal motion in the multi-dimensional space. The multidimensional structure serves as a feature representation of multidimensional variables and represents temporally related variables in such a way that temporal evolution is also modeled as a multidimensional process of continuous evolution. The multidimensional structure is computed through a novel approach of learning from multidimensional features in a set of labeled items by using a multi-layer recurrent neural network. Experiments on large-scale public datasets show that we achieve state-of-the-art performance on real-world datasets.
Towards a Semantics of Logic Program Induction, Natural Language Processing and Turing Machines
Joint Image-Visual Grounding of Temporal Memory Networks with Data-Adaptive Layerwise Regularization
Learning Discrete Dynamical Systems Based on Interacting with the Information Displacement Model
Towards a better understanding of the intrinsic value of training topic modelsWe present a nonlinear model to model the temporal evolution of human knowledge about the world. Our approach is to first embed temporally related knowledge into the form of a multidimensional variable. We then embed the inter- and intra-variable covariate into a multidimensional structure in order to model the temporal motion in the multi-dimensional space. The multidimensional structure serves as a feature representation of multidimensional variables and represents temporally related variables in such a way that temporal evolution is also modeled as a multidimensional process of continuous evolution. The multidimensional structure is computed through a novel approach of learning from multidimensional features in a set of labeled items by using a multi-layer recurrent neural network. Experiments on large-scale public datasets show that we achieve state-of-the-art performance on real-world datasets.