### Article

## Repeated games of incomplete information with large sets of states

The model of multistage insider trading between two market agents for one-type risky assets is considered. One of the players (insider) has private information about liquidation value of the asset. At each step of the bidding each player simultaneously proposes bid and ask prices for one share with fixed non-zero spread. The uninformed player uses the history of insider's moves to update his beliefs. For the bidding of unlimited duration we construct upper and lower bounds of the guaranteed insider's gain and the strategies of both players insuring these bounds. Insider's loses in the case of disclosure his private information are obtained.

For modern energy markets it is typical to use dynamic real-time pricing schemes even for residential customers. Such schemes are expected to stimulate rational energy consumption by the end customers, provide peak shaving and overall energy efficiency. But under dynamic pricing planning a household’s energy consumption becomes complicated. So automated planning of household appliances is a promising feature for developing smart home environments. Such a planning should adapt to individual user’s habits and preferences over comfort to cost balance. We propose a novel approach based on learning customer preferences expressed by a utility function. In the paper an algorithm based on inverse reinforcement learning (IRL) framework is used to infer the user’s hidden utility. We compare IRL-based approach to multiple state-of-the art machine learning techniques and the proposed earlier parametric Bayesian learning algorithm. The training and test datasets are generated by the simulated user’s behavior with different price volatility settings. The goal of the algorithms is to predict a user’s behavior based on the existing history. The IRL and Bayesian approaches showed similar performance and both of them outperforms modern machine learning algorithms such as XGBoost, random forest etc. In particular, the preference learning algorithms significantly better generalize to data generated with parameters different from the training sample. The experiments showed that preference learning approach can be especially useful for smart home automation problems where future situations can be different from those available for training.

The paper considers a game-theoretical model of bidding with asymmetric information. One player has the inside information on the liquidation price of risky asset. The model is formalized with the repeated game with incomplete information on the side of uninformed player. We consider the case of external stopping of the game at the random moment. Insider's expected profit in the game of random duration if she applies the strategy optimal in infinite-stage game is obtained. This result allows to calculate the loss of insider in case of sudden disclosure of his private information.

We consider a discrete model of insider trading in terms of repeated games with incomplete information. The solution of the bidding game of beforehand unlimited duration was obtained by V. Domansky (2007). Insider's optimal strategy in the infinite stage game generates the simple random walk of posterior probabilities over the lattice l/m, l=0,...,m with absorption at the extreme points 0 and 1 and provides the expected gain 1/2 per step to insider. In this paper we calculate insider's profit in the game of any finite duration when he applies the strategy above. It is shown that this strategy is his epsilon-optimal strategy in n-stage game, where epsilon decreases exponentially. This means that the sequence of n-stage game values converges to the value of infinite game at least exponentially. The result obtained is interpreted as the loss of insider in the case of sudden disclosure of his private information. For the special case we compare obtained insider's profit with the exact game value (result of V. Kreps, 2009) and demonstrate that error term in the case of optimal insider's behaviour also decreases exponentially.

Bayesian inference was once a gold standard for learning with neural networks, providing accurate full predictive distributions and well calibrated uncertainty. However, scaling Bayesian inference techniques to deep neural networks is challenging due to the high dimensionality of the parameter space. In this paper, we construct low-dimensional subspaces of parameter space, such as the first principal components of the stochastic gradient descent (SGD) trajectory, which contain diverse sets of high performing models. In these subspaces, we are able to apply elliptical slice sampling and variational inference, which struggle in the full parameter space. We show that Bayesian model averaging over the induced posterior in these subspaces produces accurate predictions and well-calibrated predictive uncertainty for both regression and image classification.

Topic modelling is an area of text mining that has been actively developed in the last 15 years. A probabilistic topic model extracts a set of hidden topics from a collection of text documents. It defines each topic by a probability distribution over words and describes each document with a probability distribution over topics. In applications, there are often many requirements, such as, for example, problem-specific knowledge and additional data, to be taken into account. Therefore, it is natural for topic modelling to be considered a multiobjective optimization problem. However, historically, Bayesian learning became the most popular approach for topic modelling. In the Bayesian paradigm, all requirements are formalized in terms of a probabilistic generative process. This approach is not always convenient due to some limitations and technical difficulties. In this work, we develop a non-Bayesian multiobjective approach called the Additive Regularization of Topic Models (ARTM). It is based on regularized Maximum Likelihood Estimation (MLE), and we show that many of the well-known Bayesian topic models can be re-formulated in a much simpler way using the regularization point of view. We review some of the most important types of topic models: multimodal, multilingual, temporal, hierarchical, graph-based, and short-text. The ARTM framework enables easy combination of different types of models to create new models with the desired properties for applications. This modular 'lego-style' technology for topic modelling is implemented in the open-source library BigARTM. © 2017 FRUCT.

In this article Russian stock market information efficiency is analyzed. The quantitative measure of the analysis is the indicator of Shannon entropy. On the basis of logit model the relation between the level of information efficiency and financial crisis probability is investigated.

A model for organizing cargo transportation between two node stations connected by a railway line which contains a certain number of intermediate stations is considered. The movement of cargo is in one direction. Such a situation may occur, for example, if one of the node stations is located in a region which produce raw material for manufacturing industry located in another region, and there is another node station. The organization of freight traﬃc is performed by means of a number of technologies. These technologies determine the rules for taking on cargo at the initial node station, the rules of interaction between neighboring stations, as well as the rule of distribution of cargo to the ﬁnal node stations. The process of cargo transportation is followed by the set rule of control. For such a model, one must determine possible modes of cargo transportation and describe their properties. This model is described by a ﬁnite-dimensional system of diﬀerential equations with nonlocal linear restrictions. The class of the solution satisfying nonlocal linear restrictions is extremely narrow. It results in the need for the “correct” extension of solutions of a system of diﬀerential equations to a class of quasi-solutions having the distinctive feature of gaps in a countable number of points. It was possible numerically using the Runge–Kutta method of the fourth order to build these quasi-solutions and determine their rate of growth. Let us note that in the technical plan the main complexity consisted in obtaining quasi-solutions satisfying the nonlocal linear restrictions. Furthermore, we investigated the dependence of quasi-solutions and, in particular, sizes of gaps (jumps) of solutions on a number of parameters of the model characterizing a rule of control, technologies for transportation of cargo and intensity of giving of cargo on a node station.

This proceedings publication is a compilation of selected contributions from the “Third International Conference on the Dynamics of Information Systems” which took place at the University of Florida, Gainesville, February 16–18, 2011. The purpose of this conference was to bring together scientists and engineers from industry, government, and academia in order to exchange new discoveries and results in a broad range of topics relevant to the theory and practice of dynamics of information systems. Dynamics of Information Systems: Mathematical Foundation presents state-of-the art research and is intended for graduate students and researchers interested in some of the most recent discoveries in information theory and dynamical systems. Scientists in other disciplines may also benefit from the applications of new developments to their own area of study.