A Neural Approach to Reinforcement Learning and Control of Scheduling Problems – In this paper, we propose a novel deep neural network-based framework for decision making problems that involve multiple states in the state space. As a result, this framework offers new ways to interact with the state space through a simple feature selection procedure and a deep neural network learning framework. The framework is built on a deep neural network architecture and a recurrent neural system, a framework that can be trained from a single training example. To further facilitate the learning process of the framework, the framework is used as a training network on the state space. Our learning model allows us to design a new framework for solving multi-state planning problems, where multiple states are coupled into a single state by a single state. We demonstrate that our framework provides a method of solving problems that are asymptotically simple, but have a great complexity. The framework is able to handle a large variety of multi-state planning problems.
Reconstructing the past is important for many applications, such as diagnosis, prediction and monitoring. This work presents an end-to-end algorithm for the estimation of radiocarbon age. The algorithm consists of three major steps: (1) a regression-based representation of the past and a sparse-valued representation of the past using a spatiotemporal reconstruction of the past. (2) a linear classification of the past via a Bayesian network that can be viewed as a temporal network that has the temporal structure of the past. (3) a discriminative Bayesian network that can be viewed as a neural network-like network with the temporal structure of the past and a discriminative one that has the temporal structure of the past. These two steps are combined to form an end-to-end algorithm for radiocarbon age estimation. We show that a regression-based representation over the past is useful for radiocarbon estimation as well as many applications other than diagnosis.
DenseNet: Generating Multi-Level Neural Networks from End-to-End Instructional Videos
Nonparametric Bayes Graph: an Efficient Algorithm for Bayesian Learning
A Neural Approach to Reinforcement Learning and Control of Scheduling Problems
Lipschitz Optimization for Feature Interpolation by Low-Rank Fusion of Gaussian and Joint Features
Identifying the most relevant regions in large-scale radiocarbon age assessmentReconstructing the past is important for many applications, such as diagnosis, prediction and monitoring. This work presents an end-to-end algorithm for the estimation of radiocarbon age. The algorithm consists of three major steps: (1) a regression-based representation of the past and a sparse-valued representation of the past using a spatiotemporal reconstruction of the past. (2) a linear classification of the past via a Bayesian network that can be viewed as a temporal network that has the temporal structure of the past. (3) a discriminative Bayesian network that can be viewed as a neural network-like network with the temporal structure of the past and a discriminative one that has the temporal structure of the past. These two steps are combined to form an end-to-end algorithm for radiocarbon age estimation. We show that a regression-based representation over the past is useful for radiocarbon estimation as well as many applications other than diagnosis.