In this work, we propose an automated method to identify semantic bugs in student programs, called ATAS, which builds upon the recent advances in both symbolic execution and active learning. Symbolic execution is a program analysis technique which can generate test cases through symbolic constraint solving. Our method makes use of a reference implementation of the task as its sole input. We compare our method with a symbolic execution-based baseline on 6 programming tasks retrieved from CodeForces comprising a total of 23K student submissions. We show an average improvement of over 2.5x over the baseline in terms of runtime (thus making it more suitable for online evaluation), without a significant degradation in evaluation accuracy.

Convolutional Neural Networks (CNNs) require a large amount of annotated data to learn from, which is often difficult to obtain in the medical domain. In this paper we show that the sample complexity of CNNs can be significantly improved by using 3D roto-translation group convolutions (G-Convs) instead of the more conventional translational convolutions. These 3D G-CNNs were applied to the problem of false positive reduction for pulmonary nodule detection, and proved to be substantially more effective in terms of performance, sensitivity to malignant nodules, and speed of convergence compared to a strong and comparable baseline architecture with regular convolutions, data augmentation and a similar number of parameters. For every dataset size tested, the G-CNN achieved a FROC score close to the CNN trained on ten times more data.

Existing supervised approaches didn't make use of the low-level features which are actually effective to this task. And another deficiency is that they didn't consider the relation between pixels, which means effective features are not extracted. In this paper, we proposed a novel convolutional neural network which make sufficient use of low-level features together with high-level features and involves atrous convolution to get multi-scale features which should be considered as effective features. Our model is tested on three standard benchmarks - DRIVE, STARE, and CHASE databases. The results presents that our model significantly outperforms existing approaches in terms of accuracy, sensitivity, specificity, the area under the ROC curve and the highest prediction speed. Our work provides evidence of the power of wide and deep neural networks in retinal blood vessels segmentation task which could be applied on other medical images tasks.

Computer vision systems for automatic image categorization have become accurate and reliable enough that they can run continuously for days or even years as components of real-world commercial applications. A major open problem in this context, however, is quality control. Good classification performance can only be expected if systems run under the specific conditions, in particular data distributions, that they were trained for. Surprisingly, none of the currently used deep network architectures has a built-in functionality that could detect if a network operates on data from a distribution that it was not trained for and potentially trigger a warning to the human users. In this work, we describe KS(conf), a procedure for detecting such outside of the specifications operation. Building on statistical insights, its main step is the applications of a classical Kolmogorov-Smirnov test to the distribution of predicted confidence values. We show by extensive experiments using ImageNet, AwA2 and DAVIS data on a variety of ConvNets architectures that KS(conf) reliably detects out-of-specs situations. It furthermore has a number of properties that make it an excellent candidate for practical deployment: it is easy to implement, adds almost no overhead to the system, works with all networks, including pretrained ones, and requires no a priori knowledge about how the data distribution could change. 1 Introduction With the emergence of deep convolutional networks (ConvNets), computer vision systems have become accurate and reliable enough to perform tasks of practical relevance autonomously and reliably over long periods of time. This work was in parts funded by the European Research Council under the European Unions Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no 308036. C. H. Lampert, IST Austria, Email: chl@ist.ac.at 2 Rémy Sun, Christoph H. Lampert "ski" "shovel" "web site" "tennis ball" Figure 1 Illustration of within specification and outside of specifications behavior of a Conv-Net (here: VGG19, trained on ILSVRC2012). Left image: prediction on images that the network was trained to recognize. We observe: a standard multi-class network always predicts one of its predefined class labels, even if the current input is distorted, or even completely different, from what it was trained for. A major concern in our society about automatic decision systems is their reliability: if decisions are made by a trained classifier instead of a person, how can we be sure that the system works reliably now, and that it will continue to do so in the future?

However, many objectives of practical interest are more complex than simply average loss. Examples include balancing performance or loss with fairness across people, as well as balancing precision and recall. We prove that, from a computational perspective, fairly general families of complex objectives are not significantly harder to optimize than standard averages, by providing polynomial-time reductions, i.e., algorithms that optimize complex objectives using linear optimizers. The families of objectives included are arbitrary continuous functions of average group performances and also convex objectives. We illustrate with applications to fair machine learning, fair optimization and F1-scores.

We develop an algorithm for minimizing a function using $n$ batched function value measurements at each of $T$ rounds by using classifiers to identify a function's sublevel set. We show that sufficiently accurate classifiers can achieve linear convergence rates, and show that the convergence rate is tied to the difficulty of active learning sublevel sets. Further, we show that the bootstrap is a computationally efficient approximation to the necessary classification scheme. The end result is a computationally efficient derivative-free algorithm requiring no tuning that consistently outperforms other approaches on simulations, standard benchmarks, real-world DNA binding optimization, and airfoil design problems whenever batched function queries are natural.

Accurately predicting the possible behaviors of traffic participants is an essential capability for future autonomous vehicles. The majority of current researches fix the number of driving intentions by considering only a specific scenario. However, distinct driving environments usually contain various possible driving maneuvers. Therefore, a intention prediction method that can adapt to different traffic scenarios is needed. To further improve the overall vehicle prediction performance, motion information is usually incorporated with classified intentions. As suggested in some literature, the methods that directly predict possible goal locations can achieve better performance for long-term motion prediction than other approaches due to their automatic incorporation of environment constraints. Moreover, by obtaining the temporal information of the predicted destinations, the optimal trajectories for predicted vehicles as well as the desirable path for ego autonomous vehicle could be easily generated. In this paper, we propose a Semantic-based Intention and Motion Prediction (SIMP) method, which can be adapted to any driving scenarios by using semantic-defined vehicle behaviors. It utilizes a probabilistic framework based on deep neural network to estimate the intentions, final locations, and the corresponding time information for surrounding vehicles. An exemplar real-world scenario was used to implement and examine the proposed method.

Information retrieval and binary classification can be considered equivalent problems in principle. Information retrieval means to mark documents in a set of candidate documents as relevant or non-relevant for some question, on the basis of the properties of the documents. For binary classification, the problem is to distinguish between the'positive' and'negative' instances from a dataset, based on the features of the instances. Hence, from an abstract point of view, information retrieval is a special case of binary classification, with the documents being instances, the document properties being features and'relevant' being translated as'positive'. In practice, however, the general concepts from binary classification are not always helpful for information retrieval applications. The fact that often the proportion of relevant documents in a set of documents subject to a search is small or even very small is only one of the reasons for information retrieval to be considered a field of research for its own. As a consequence, some performance measures for information retrieval methods differ from those in use for binary classifiers or are called by different names. Precision and recall are possibly the most popular performance measures(see Chapter 8 of Manning et al., 2008, for a list of performance measures) for information retrieval methods: - Precision is the proportion of documents (instances) that are truly relevant (positive) among those documents which have been predicted relevant (positive). The term precision is also commonly used (with the same meaning) in binary classification.

Layers are Python functions that take Tensors and configuration options as input and produce other tensors as output. Once the necessary Tensors have been composed, the user can convert the result into an Estimator via a model function.

WrestleMania 34 is WWE's biggest event of 2018 in just about every way possible. It's set to have the year's largest attendance with more matches than any other show, and it'll last much longer than the other pay-per-views of the past 12 months. None of that, however, necessarily adds up to an expensive price tag for viewers Sunday night. WrestleMania 34 will cost $54.99 to order on pay-per-view, but there is an easy way to watch it with a free live stream. WrestleMania 34 will be broadcast live on the WWE Network, just like every other pay-per-view.