This paper is discusses the problems of the short-term forecasting of financial time series using supervised machine learning (ML) approach. For this goal, we applied several the most powerful methods including Support Vector Machine (SVM), Multilayer Perceptron (MLP), Random Forests (RF) and Stochastic Gradient Boosting Machine (SGBM). As dataset were selected the daily close prices of two stock index: SP 500 and NASDAQ, two the most capitalized cryptocurrencies: Bitcoin (BTC), Ethereum (ETH), and exchange rate of EUR-USD. As features we used only the past price information. To check the efficiency of these models we made out-of-sample forecast for selected time series by using one step ahead technique. The accuracy rates of the forecasted prices by using ML models were calculated. The results verify the applicability of the ML approach for the forecasting of financial time series. The best out of sample accuracy of short-term prediction daily close prices for selected time series obtained by SGBM and MLP in terms of Mean Absolute Percentage Error (MAPE) was within 0.46-3.71 %. Our results are comparable with accuracy obtained by Deep learning approaches.
Abstract The paper deals with the architecture and design of the complex distributed classifier for the intelligent video surveillance systems considering the contemporary tendency to detect the abnormal or suspicious behavior of the individuals by means of behavioral features set analysis. This paper focuses on the implementation of multiagent systems concept and the distributed ledger technology to the distributed message board architecture. Two selected approaches to the distributed ledger implementation are analyzed and estimated in terms of classifiers cooperation. Some simulation results are provided and discussed in terms of time consumption.
Clément Bénard, Gérard Biau, Sébastien da Veiga, Erwan Scornet
We introduce SIRUS (Stable and Interpretable RUle Set) for regression, a stable rule learning algorithm which takes the form of a short and simple list of rules. State-of-the-art learning algorithms are often referred to as "black boxes" because of the high number of operations involved in their prediction process. Despite their powerful predictivity, this lack of interpretability may be highly restrictive for applications with critical decisions at stake. On the other hand, algorithms with a simple structure-typically decision trees, rule algorithms, or sparse linear models-are well known for their instability. This undesirable feature makes the conclusions of the data analysis unreliable and turns out to be a strong operational limitation. This motivates the design of SIRUS, which combines a simple structure with a remarkable stable behavior when data is perturbed. The algorithm is based on random forests, the predictive accuracy of which is preserved. We demonstrate the efficiency of the method both empirically (through experiments) and theoretically (with the proof of its asymptotic stability). Our R/C++ software implementation sirus is available from CRAN.
Alexander Loktyushin, Kai Herz, Nhan Tam Dang, Felix Glang · 10 authors
Purpose: A supervised learning framework is proposed to automatically generate MR sequences and corresponding reconstruction without human knowledge on MR strategies. This enables a target-based optimization from scratch, as well as exploration of novel and flexible MR sequence strategies. Methods: The entire scanning and reconstruction process is simulated end-to-end in terms of RF events, gradient moment events in x and y, and delay times, acting on the input model spin system given in terms of proton density, T1 and T2, and B0. As proof of concept we use both conventional MR images but also binary masks and T1 maps as a target and optimize from scratch using the loss defined by data fidelity, SAR, and scan time. Results: In a first attempt, MRzero learns all gradient and RF events from zero, and is able to generate the aimed at target image. Appending a neural network layer to the reconstruction module also arbitrary targets were learned successfully. Experiments could be translated to image acquisition at a real system (3T Siemens, PRISMA) and could be verified in measurements of phantoms and the human brain in vivo. Discussion/Conclusion: We have developed a fully automated MR sequence generator based on Bloch equation simulations and supervised learning. While we focus on the method herein, having such a differentiable digital MR twin at hand paves the way to a novel way of generating MR sequence and reconstruction solely governed by the target provided, which can be a certain MR contrast, but the possibilities for targets are limitless, e.g. quantification, segmentation, as well as contrasts of other image modalities.
zkSNARKS can be described as zero-knowledge: No secret information is revealed by the proof; Succinct: The size of the proof that is generated is small; Non-interactive: no challenge-response protocol; and ARgument of Knowledge: It is computationally intractable for the prover to produce a fake proof.
Artificial neural networks are modern methods suitable for solving the problem of nonlinear dependency approximation, which is successfully applied in many fields. This paper compares the predictive capabilities of Back Propagation, Radial Basis Function, Extreme Learning Machine, and Long-Short Term Memory neural networks to determine which artificial intelligence algorithm is best for modeling the price of Bitcoin opening. The criterion for comparing network performance was the standard deviation, the mean absolute deviation, and the accuracy of predicting the direction of change of course. At the same time, in the study of time series, it is recommended to perform a comprehensive data analysis using appropriate networks, depending on the length of the series and the specificity of the database.
In this paper, we apply Long Short-Term Memory (LSTM) neural networks to model the token price time series data which incorporate the local topological measures of investor transaction network and market summaries. In addition, we propose a novel LSTM-based model using the leave-one-out cross-validation technique and utilizing the network motif analysis. The numerical results show that the proposed LSTM-based model could significantly improve the performance of the prediction of token price compared to the benchmark LSTM-based models and deep portfolios regardless of the training and testing data split ratio. Some concluding remarks and future research directions are provided.
А. Е. Мисник, S. K. Krutalevich, Siarhei А. Prakapenka, Peter Borovykh · 5 authors
Neural network is the universal approximator, but its precision highly depends on sufficient set of inputs. Cryptocurrencies have great volatility, due to absence of fundamentals to back up their price. In this paper we analyze approaches to obtain additional parameters for neural networks and explore their impact on its prediction accuracy. This study indicates significant improvement of neural network predictions due to inclusion of wider selection of relevant data points.
Nor Azizah Hitam, Amelia Ritahani Ismail, Faisal Saeed
Forecasting accurate future price is very important in financial sector. An optimized Support Vector Machine (SVM) based on Particle Swarm Optimization (PSO) is introduced in forecasting the cryptocurrency future price. It is part of Artificial Intelligence (AI) that uses previous experience to forecast future price. Analysts and investors generally combine fundamental and technical analysis prior to decide the best price to execute their trades. Some may use Machine Learning Algorithms to execute their trades. However, forecasting result using basic SVM algorithms does not really promising. On the other hands, Particle Swarm Optimization (PSO) is known as a better algorithm for a static and simple optimization problem. Therefore, PSO is introduced to optimize the algorithms of SVM in cryptocurrency forecasting. The experiment of selected cryptocurrencies is conducted for this classifier. The experimental result demonstrates that an optimized SVM-PSO algorithm can effectively forecast the future price of cryptocurrency thus outperforms the single SVM algorithms.
Distributed processing and control are critical to supports distributed intelligence and autonomy of multi-degree-of-freedom motion systems. Measurements and fusion of spatiotemporal physical quantities imply data-intensive computing and spatial distribution of computing resources to enable control and processing of large data sets from image and inertial sensors. We examine distributed and asynchronous processing nodes which process information independently deriving partial solutions. There are multiple sensing-and-processing nodes in each individual agent. Each node comprises solid-state or MEMS multi-degree-of-freedom sensors with ASICs which process and fuse data. On-node computing supports distributed processing. Adaptive bottom-up organization ensures data aggregation and data management with operation on sub-samples or hashed sets of large source datasets. Cooperative distributed processing is essential in centralized, decentralized and behavioral coordination. In the centralized organization, a central processor may not ensure adequacy. A network of semi-autonomous on-device processing sensors may interact to solve specific tasks and validate solutions. Problem allocation, partitioning, coordination and other tasks are implemented using software-and hardware-supported algorithms and protocols. This paper contributes to design of next generation of systems with distributed multi-node processing capabilities.
Target Tracking and Data Fusion in Sensor Networks
This paper explores the relationship between the features of Bitcoin and the next day change in the price of Bitcoin using an Artificial Neural Network ensemble approach called Genetic Algorithm based Selective Neural Network Ensemble, constructed using Multi-Layered Perceptron as the base model for each of the neural network in the ensemble. To better understand the practicality and its effectiveness in real-world application, the ensemble was used to predict the next day direction of the price of Bitcoin given a set of approximately 200 features of the cryptocurrency over a span of 2 years. Over a span of 50 days, a trading strategy based on the ensemble was compared against a “previous day trend following” trading strategy through back-testing. The former trading strategy generated almost 85% returns, outperforming the “previous day trend following” trading strategy which produced an approximate 38% returns and a trading strategy that follows the single, best MLP model in the ensemble that generated approximately 53% in returns.
From a visual standpoint it is often easy to point out whether a system is considered to be self-organizing or not, though a quantitative approach would be more helpful. Information theory, as introduced by Shannon, provides the right tools not only quantify self-organization, but also to investigate it in relation to the information processing performed by individual agents within a collective. This thesis sets out to introduce methods to quantify spatial self-organization in collective systems in the continuous domain as a means to investigate morphogenetic processes. In biology, morphogenesis denotes the development of shapes and form, for example embryos, organs or limbs. Here, I will introduce methods to quantitatively investigate shape formation in stochastic particle systems. In living organisms, self-organization, like the development of an embryo, is a guided process, predetermined by the genetic code, but executed in an autonomous decentralized fashion. Information is processed by the individual agents (e.g. cells) engaged in this process. Hence, information theory can be deployed to study such processes and connect self-organization and information processing. The existing concepts of observer based self-organization and relevant information will be used to devise a framework for the investigation of guided spatial self-organization. Furthermore, local information transfer plays an important role for processes of self-organization. In this context, the concept of synergy has been getting a lot attention lately. Synergy is a formalization of the idea that for some systems the whole is more than the sum of its parts and it is assumed that it plays an important role in self-organization, learning and decision making processes. In this thesis, a novel measure of synergy will be introduced, that addresses some of the theoretical problems that earlier approaches posed.
Information theory is a powerful tool to express principles to drive autonomous systems because it is domain invariant and allows for an intuitive interpretation. This paper studies the use of the predictive information (PI), also called excess entropy or effective measure complexity, of the sensorimotor process as a driving force to generate behavior. We study nonlinear and nonstationary systems and introduce the time-local predicting information (TiPI) which allows us to derive exact results together with explicit update rules for the parameters of the controller in the dynamical systems framework. In this way the information principle, formulated at the level of behavior, is translated to the dynamics of the synapses. We underpin our results with a number of case studies with high-dimensional robotic systems. We show the spontaneous cooperativity in a complex physical system with decentralized control. Moreover, a jointly controlled humanoid robot develops a high behavioral variety depending on its physics and the environment it is dynamically embedded into. The behavior can be decomposed into a succession of low-dimensional modes that increasingly explore the behavior space. This is a promising way to avoid the curse of dimensionality which hinders learning systems to scale well.
We present an algorithm, inspired by self-organization and stigmergy observed in biological swarms, for managing multiple sensors tracking large numbers of targets. We devise a decentralized architecture wherein autonomous sensors manage their own data collection resources and task themselves. Sensors cannot communicate with each other directly; however, a global track file, which is continuously broadcast, allows the sensors to infer their contributions to the global estimation of target states. Sensors can transmit their data (either as raw measurements or some compressed format) only to a central processor where their data are combined to update the global track file. We outline information-theoretic rules for the general multiple-sensor Bayesian target tracking problem. We provide specific formulas for problems dominated by additive white Gaussian noise. Using Cramér-Rao lower bounds as surrogates for error covariances, we illustrate, using numerical scenarios involving ballistic targets, that the bioinspired algorithm is highly scalable and performs very well for large numbers of targets.
Target Tracking and Data Fusion in Sensor Networks
Distributed Sensor Networks and Detection Algorithms