Learn More
The recently introduced Deep Q-Networks (DQN) algorithm has gained attention as one of the first successful combinations of deep neural networks and reinforcement learning. Its promise was demonstrated in the Arcade Learning Environment (ALE), a challenging framework composed of dozens of Atari 2600 games used to evaluate general competency in AI. It(More)
Activity-dependent structural remodeling is an important aspect of neuronal plasticity. In the previous researches, neuronal structure variations resulting from external interventions were detected by the imaging instruments such as the fluorescence microscopy, the scanning/transmission electron microscopy (SEM/TEM) and the laser confocal microscopy. In(More)
The probabilistic sentential decision diagram (PSDD) was recently introduced as a tractable representation of probability distributions that are subject to logical constraints. Meanwhile, efforts in tractable learning achieved great success inducing complex joint distributions from data without constraints, while guaranteeing efficient exact probabilistic(More)
Electrophysiology and mechanics are two essential components in the functions of cardiomyocytes and skeletal muscle cells. The simultaneous recording of electrophysiological and mechanical activities is important for the understanding of mechanisms underlying cell functions. For example, on the one hand, mechanisms under cardiovascular drug effects will be(More)
A novel method to estimate actual size of underwater moving objects on condition of no any references was proposed. Defocusing effect resulted from different distance between moving objects and lens was analyzed. Blurred edge width in out-of-focus image was detected by using wavelet transformation. The relations between blurred edge width and real distance(More)
Cell patterning has been widely used in research on fundamental cell biology and in applications such as tissue engineering, neuron network formation, cell based biosensor and drug screening. Although various methods have been developed, cell patterning in an enclosed microfluidic device at single cell level remains challenging. This paper describes a(More)
To carry out image post-processing efficiently, it is necessary representing image with statistics principle in order to increase researchers' prior knowledge to image. In the essay, a representation about special color fundus images is proposed by extracting corresponding image color space's component grayscale. And then, the qualitative and quantitative(More)
In medicine, whether the fundus image is clear or not would affect the doctor's diagnosis of the disease. In this paper, the feature vector of clear or blurred image is constructed by the fundus image edge degree and the transformed energy of the wavelet coefficients. The fisher criterion is adopted in order to determine the optimal projection direction and(More)
In the process of grain storage, there are many losses of grain quantity and quality for the sake of insects. As a result, it is necessary to find a rapid and economical method for detecting insects in the grain. The paper innovatively proposes a model of detecting internal infestation in wheat by combining pattern recognition and BioPhoton Analytical(More)
Probabilistic sentential decision diagrams (PSDDs) were recently introduced as a tractable and interpretable representation of discrete probability distributions. PSDDs are tractable because they support a wide range of queries efficiently. They are interpretable because each parameter in the PSDD represents a conditional probability, as in Bayesian(More)