Semi-supervised learning is a machine learning approach that tackles the challenge of having a large set of unlabeled data and few labeled ones. In this paper we adopt a semi-supervised self-training method to increase the amount of training data, prevent overfitting and improve the performance of deep models by proposing a novel selection algorithm that prevents mistake reinforcement which is a common thing in conventional self-training models. The model leverages, unlabeled data and specifically, after each training, we first generate pseudo-labels on the unlabeled set to be added to the labeled training samples. Next, we select the top-k most-confident pseudo-labeled images from each unlabeled class with their pseudo-labels and update the training data, and retrain the network on the updated training data. The method improves the accuracy in two-fold; bridging the gap in the appearance of visual objects, and enlarging the training set to meet the demands of deep models. We demonstrated the effectiveness of the model by conducting experiments on four state-of-the-art fine-grained datasets, which include Stanford Dogs, Stanford Cars, 102-Oxford flowers, and CUB-200-2011. We further evaluated the model on some coarsegrain data. Experimental results clearly show that our proposed framework has better performance than some previous works on the same data; the model obtained higher classification accuracy than most of the supervised learning models.

Traffic sign recognition is a classification problem that poses challenges for computer vision and machine learning algorithms. Although both computer vision and machine learning techniques have constantly been improved to solve this problem, the sudden rise in the number of unlabeled traffic signs has become even more challenging. Large data collation and labeling are tedious and expensive tasks that demand much time, expert knowledge, and fiscal resources to satisfy the hunger of deep neural networks. Aside from that, the problem of having unbalanced data also poses a greater challenge to computer vision and machine learning algorithms to achieve better performance. These problems raise the need to develop algorithms that can fully exploit a large amount of unlabeled data, use a small amount of labeled samples, and be robust to data imbalance to build an efficient and high-quality classifier. In this work, we propose a novel semi-supervised classification technique that is robust to small and unbalanced data. The framework integrates weakly-supervised learning and self-training with self-paced learning to generate attention maps to augment the training set and utilizes a novel pseudo-label generation and selection algorithm to generate and select pseudo-labeled samples. The method improves the performance by: (1) normalizing the class-wise confidence levels to prevent the model from ignoring hard-to-learn samples, thereby solving the imbalanced data problem; (2) jointly learning a model and optimizing pseudo-labels generated on unlabeled data; and (3) enlarging the training set to satisfy the hunger of deep learning models. Extensive evaluations on two public traffic sign recognition datasets demonstrate the effectiveness of the proposed technique and provide a potential solution for practical applications.

With the increasing complexity of maritime traffic, vessel trajectory prediction becomes a vital part of the maritime security. Although many methods working on this problem have been proposed including machine learning methods, physical model methods and deep learning methods. But all these methods can not take long-time prediction and multidimensional prediction in to account at the same time. Besides, even some methods avoided the curved trajectory prediction deliberately. In this paper we investigated a novel approach based on Variational LSTM to predict the vessel trajectory using Automatic Identification System (AIS) data. The predicted time scope in ranges from about 5 minutes to about 20 minutes and the predicted features including time, longitude, latitude, speed over ground (SOG) and course over ground (COG). To avoid damaging the structure of the raw AIS data, we have not used any interpolation method when processed it. Instead, we calculated the time increment between two continuous data points. Then we used a novel method based on time increment to select out those trajectory segments with preferable time continuity to make dataset. Lastly, in long-time and multidimensional prediction experiments, variational LSTM achieved the better results and showed the potential in the long-time trajectory prediction comparing with the LSTM. In 20-time-steps position prediction, the MSE of variational LSTM was only 6.0303e-4 but that value of LSTM reached to 6.0014e-3.

Deep Neural Networks have provided excellent performance in a variety of applications. However, their superior performance comes with a huge expense of collecting a correctly-annotated large-scale training set, and a little impracticality of preparing such a training set in many applications. In this work, we study another natural type of weak supervision, complementary-label learning, to address this problem. Complementary-label learning refers to train the Deep Neural Networks by the usage of only complementary labels, and a complementary label indicates one of the classes that the sample does not belong to. This paper first presents a general risk formulation for complementary label learning through an adoption of arbitrary losses designed for ordinary-label learning. We then theoretically analyze that our method is applicable for any loss functions to learn deep neural networks with complementary labels in the framework of risk minimization. Experimental results on different benchmark datasets demonstrate that our approach outperforms current state-of-the-art methods.

This paper presents a new compact canonical-based algorithm to solve the problem of singleoutput completely specified NPN Boolean matching.We propose a new signature vector Boolean difference and cofactor (DC) signature vector. Our algorithm utilizes the Boolean difference, cofactor signature and symmetry properties to search for canonical transformations. The use of symmetry and Boolean difference notably reduces the search space and speeds up the Boolean matching process compared to the algorithm proposed in [1]. We tested our algorithm on a large number of circuits. The experimental results showed that the average runtime of our algorithm 37% higher and its average search space 67% smaller compared to [1] when tested on general circuits.

In this paper, we address an NPN Boolean matching algorithm. The proposed structural difference signature (SDS) of a Boolean function significantly reduces the search space in the Boolean matching process. The paper analyses the size of the search space from three perspectives: the total number of possible transformations, the number of candidate transformations and the number of decompositions. We test the search space and run time on a large number of randomly generated circuits and Microelectronics Center of North Carolina (MCNC) benchmark circuits with 7–22 inputs. The experimental results show that the search space of Boolean matching is greatly reduced and the matching speed is obviously accelerated.