Schedae Informaticae is an international journal dedicated to computer science and its mathematical foundations. The journal publishes high quality, refereed research papers dealing with the following topics: theoretical computer science, computer science applications, mathematical foundations of computer science, applications of mathematical methods to computer science, computational and numerical mathematics.

Infinite length of trajectories is an almost universal assumption in the theoretical foundations of reinforcement learning. In practice learning occurs on finite trajectories. In this paper we examine a specific result of this disparity, namely a strong bias of the time-bounded Every-visit Monte Carlo value estimator. This manifests as a vastly different learning dynamic for algorithms that use value predictors, including encouraging or discouraging exploration. We investigate these claims theoretically for a one dimensional random walk, and empirically on a number of simple environments. We use GAE as an algorithm involving a value predictor and evolution strategies as a reference point.

We investigate performance of a gradient descent optimization (GR) applied to the traffic signal setting problem and compare it to genetic algorithms. We used neural networks as metamodels evaluating quality of signal settings and discovered that both optimization methods produce similar results, e.g., in both cases the accuracy of neural networks close to local optima depends on an activation function (e.g., TANH activation makes optimization process converge to different minima than ReLU activation).

We present a novel modification of context encoder loss function, which results in more accurate and plausible inpainting. For this purpose, we introduce gradient attention loss component of loss function, to suppress the common problem of inconsistency in shapes and edges between the inpainted region and its context. To this end, the mean absolute error is computed not only for the input and output images, but also for their derivatives. Therefore, model concentrates on areas with larger gradient, which are crucial for accurate reconstruction. The positive effects on inpainting results are observed both for fully-connected and fully-convolutional models tested on MNIST and CelebA datasets.

One of the most important optimization tasks in the industry at the current time is the object packing problem. Although several methods have been developed for the purpose of solving it, they are usually only able to optimize placement locally and as such are heavily dependent on the choice of the initial setting -hence the need for trying out multiple possible starting points, which impacts algorithm running time. In this paper we present a neural networkbased model which provides sensible starting points in a linear time.

In this paper, we study the feasibility of using a neural network to learn a fitness function for a machine translation system based on a genetic algorithm termed GAMaT. The neural network is learned on features extracted from pairs of source sentences and their translations. The fitness function is trained in order to estimate the BLEU of a translation as precisely as possible. The estimator has been trained on a corpus of more than 1.3 million data. The performance is very promising: the difference between the real BLEU and the one given by the estimator is equal to 0.12 in terms of Mean Absolute Error. Keywords: Statistical Machine Translation, Genetic algorithm, Quality estimation, Neural network

For a positive integer n, n-expandable deep pushdown automata always contain no more than n occurrences of non-input symbols in their pushdowns during any computation. As its main result, the present paper demonstrates that these automata are as powerful as the same automata with only two non-input pushdown symbols-$ and #, where # always appears solely as the pushdown bottom. The paper demonstrates an infinite hierarchy of language families that follows from this main result. The paper also points out that if # is the only non-input symbol in these automata, then they characterize the family of regular languages. In its conclusion, the paper suggests open problems and topics for the future investigation.

Impact of Clustering Parameters on the Efficiency of the Knowledge Mining Process in Rule-based Knowledge Bases

Schedae Informaticae » 2017 » Volume 26 » Regression SVM for Incomplete Data A A A

Schedae Informaticae » 2016 » Volume 25 » Optimization of ℓp-regularized Linear Models via Coordinate Descent A A A