Filters








4,894 Hits in 5.9 sec

Towards realistic market simulations

Andrea Coletta, Matteo Prata, Michele Conti, Emanuele Mercanti, Novella Bartolini, Aymeric Moulin, Svitlana Vyetrenko, Tucker Balch
2021 Proceedings of the Second ACM International Conference on AI in Finance  
To addresses this challenge we propose a synthetic market generator based on Conditional Generative Adversarial Networks (CGANs) trained on real aggregate-level historical data.  ...  Backtesting, a widely used approach, consists of simulating experimental strategies while replaying historical market scenarios.  ...  We study the following stylized facts: • Generative Adversarial Networks (GANs).  ... 
doi:10.1145/3490354.3494411 fatcat:gugdew4qpza6xkxer4yelllype

Towards Realistic Market Simulations: a Generative Adversarial Networks Approach [article]

Andrea Coletta, Matteo Prata, Michele Conti, Emanuele Mercanti, Novella Bartolini, Aymeric Moulin, Svitlana Vyetrenko, Tucker Balch
2021 arXiv   pre-print
To addresses this challenge we propose a synthetic market generator based on Conditional Generative Adversarial Networks (CGANs) trained on real aggregate-level historical data.  ...  Backtesting, a widely used approach, consists of simulating experimental strategies while replaying historical market scenarios.  ...  Generative Adversarial Networks (GANs).  ... 
arXiv:2110.13287v1 fatcat:mpinu24movdnhlibaefljyyrvm

Modern ICT Network Simulator for Co-Simulations in Smart Grid Applications

Fabian Niehaus, Bastian Fraune, Giacomo Gritzan, Richard Sethmann
2022 International Conference on Cyber Warfare and Security (ICIW)  
It also demonstrates our simulators capabilities to perform real-world ICT attacks on realistic network topologies.  ...  The current transformation of power grids towards smart grids and the associated increase in the use of ICT technologies lead to an increased attack surface via the communication network.  ...  New approaches and concepts can thus be tested as in a realistic environment, even before the rollout on actual hardware.  ... 
doi:10.34190/iccws.17.1.44 fatcat:qatjwuqz7bgnteaehraynowzmi

Training robust anomaly detection using ML-Enhanced simulations [article]

Philip Feldman
2020 arXiv   pre-print
Our approach enhances simulations using neural networks trained on real-world data to create outputs that are more realistic and variable than traditional simulations.  ...  This paper describes the use of neural networks to enhance simulations for subsequent training of anomaly-detection systems.  ...  Noise Training and Generation In this approach, noise is trained independently using Generative Adversarial Neural Networks (GANs) Generative adversarial learning is a technique where a generative network  ... 
arXiv:2008.12082v2 fatcat:gapu4zh3gnhzzdjfsgplb4obsu

Systematic review of features for co‐simulating security incidents in Cyber‐Physical Systems

Ricardo M. Czekster, Charles Morisset, John A. Clark, Sadegh Soudjani, Charalampos Patsios, Peter Davison
2021 Security and Privacy  
Addressing security requires developing modeling and simulation tools that approximate and replicate adversarial behavior in the SG.  ...  The technique of composing two models of computation in a global simulation of these coupled systems is called co-simulation.  ...  The authors implemented a Smart Grid Simulator for experimentation and validation of the approach in a realistic case study.  ... 
doi:10.1002/spy2.150 fatcat:mcrafuc2tnbqnmsywuniygcbcq

Learning who is in the market from time series: market participant discovery through adversarial calibration of multi-agent simulators [article]

Victor Storchan, Svitlana Vyetrenko, Tucker Balch
2021 arXiv   pre-print
within an optimization framework to tune parameters of a simulator model with known agent archetypes to represent a market scenario.  ...  In this paper, we address the problem of multi-agent simulator parameter calibration to allow simulator capture characteristics of different market regimes.  ...  Figure 9 : 9 Generative adversarial network architecture with self-attention. Figure 10 : 10 Generative adversarial network architecture without self-attention (used for the ablation study).  ... 
arXiv:2108.00664v1 fatcat:xeilvnj7afhqhhvd4peffsjgfy

Privacy assessment in vehicular networks using simulation

Isabel Wagner, David Eckhoff
2014 Proceedings of the Winter Simulation Conference 2014  
While vehicular network algorithms are usually evaluated by means of simulation, it is a non-trivial task to assess the performance of a privacy protection mechanism.  ...  Vehicular networks are envisioned to play an important role in the building of intelligent transportation systems.  ...  We conducted a systematic literature review concerning the assessment of privacy in simulations of vehicular networks.  ... 
doi:10.1109/wsc.2014.7020152 dblp:conf/wsc/WagnerE14 fatcat:mmmlfpuf7jcd7p6p63f3ysmu6i

MASS: Mobile Autonomous Station Simulation [article]

Thomas Sandholm, Sayandev Mukherjee
2021 arXiv   pre-print
Traces are generated by a user- and context-aware trained generative adversarial network (GAN).  ...  Our results show that we beat both traditional statistical distribution fitting approaches as well as a state-of-the-art GAN time series generator across these metrics.  ...  Adversarial architecture of MassGAN Like all GAN models, MassGAN also has a generator network and a discriminator network.  ... 
arXiv:2111.09161v1 fatcat:3hkbicxavff57hoh4hzviyjury

Modeling multiple communities of interest for interactive simulation and gaming: the dynamic adversarial gaming algorithm project

Eugene Santos, Jr., Qunhua Zhao, Felicia Pratto, Adam R. Pearson, Bruce McQueary, Andy Breeden, Lee Krause, Kevin Schum, Dawn A. Trevisani
2007 Modeling and Simulation for Military Operations II  
The Dynamic Adversarial Gaming Algorithm (DAGA) project aims to provide a wargaming environment for automation of simulating dynamics of geopolitical crisis and eventually be applied to military simulation  ...  In this paper, we describe our COI modeling, the development of cultural networks, the interaction architecture, and a prototype of DAGA.  ...  A realistic adversarial simulation beyond simply modeling attrition principles requires a system capability for predicting with continual assessment of the intentions and courses of actions of all parties  ... 
doi:10.1117/12.722319 fatcat:avz6sxt5q5eerlxqc2stxy36aq

A Data-driven Market Simulator for Small Data Environments [article]

Hans Bühler, Blanka Horvath, Terry Lyons, Imanol Perez Arribas, Ben Wood
2020 arXiv   pre-print
We also contrast some classical approaches of market simulation with simulation based on generative modelling and highlight some advantages and pitfalls of the new approach.  ...  Neural network based data-driven market simulation unveils a new and flexible way of modelling financial time series without imposing assumptions on the underlying stochastic dynamics.  ...  The emergence of DNN-based financial applications is one of the driving factors that directed the interest towards highly realistic market simulators: A key factor for training these deep networks to a  ... 
arXiv:2006.14498v1 fatcat:imtmblo64jgr5gp2xfcbneac3y

Simulation-based Adversarial Test Generation for Autonomous Vehicles with Machine Learning Components [article]

Cumhur Erkan Tuncali, Georgios Fainekos, Hisahiro Ito, James Kapinski
2019 arXiv   pre-print
We present a testing framework that is compatible with test case generation and automatic falsification methods, which are used to evaluate cyber-physical systems.  ...  One of the main challenges is that many autonomous driving systems have machine learning components, such as deep neural networks, for which formal properties are difficult to characterize.  ...  Under this category, we could also potentially include generative adversarial networks [13] .  ... 
arXiv:1804.06760v4 fatcat:jxy5mrzrjjhvferrzm7jltncne

An Attack Simulation and Evidence Chains Generation Model for Critical Information Infrastructures

Eleni-Maria Kalogeraki, Spyridon Papastergiou, Themis Panayiotopoulos
2022 Electronics  
., log files, which are evidence of cyber-attacks on a system or network.  ...  This paper proposes an attack simulation and evidence chains generation model which computes all possible attack paths associated with specific, confirmed security events.  ...  simulation and evidence chains generation approach.  ... 
doi:10.3390/electronics11030404 fatcat:fs4o3ruzezg5npw5ijmy2uucci

Technological Heterogeneity and Path Diversity in Smart Home Resilience: A Simulation Approach

Amir Modarresi, John Symons
2020 Procedia Computer Science  
Our approach involves simulating and modeling the interaction of diverse smart home technologies in the context of their relationship to the core internet.  ...  Our approach involves simulating and modeling the interaction of diverse smart home technologies in the context of their relationship to the core internet.  ...  In order to do that, we generate a number of flows from various nodes in the network toward Server/Cloud node.  ... 
doi:10.1016/j.procs.2020.03.023 fatcat:iijpxfixvrgnlajxf3b6tfffmm

Data-Centric Engineering: integrating simulation, machine learning and statistics. Challenges and Opportunities [article]

Indranil Pan, Lachlan Mason, Omar Matar
2021 arXiv   pre-print
New hybrid, data-centric engineering approaches, leveraging the best of both worlds and integrating both simulations and data, are emerging as a powerful tool with a transformative impact on the physical  ...  Mechanistic models, based on physical equations, and purely data-driven statistical approaches represent two ends of the modelling spectrum.  ...  Generative modelling and simulations Deep generative models, which include variational autoencoders and generative adversarial networks (GANs) have been hugely successful in generating realistic synthetic  ... 
arXiv:2111.06223v2 fatcat:ejwhon6envhandje4twvatgliq

Sim4CV: A Photo-Realistic Simulator for Computer Vision Applications

Matthias Müller, Vincent Casser, Jean Lahoud, Neil Smith, Bernard Ghanem
2018 International Journal of Computer Vision  
We present a photo-realistic training and evaluation simulator (Sim4CV) with extensive applications across various fields of computer vision.  ...  The simulator fully integrates both several state-of-the-art tracking algorithms with a benchmark evaluation tool and a deep neural network (DNN) architecture for training vehicles to drive autonomously  ...  The differences in appearance between the simulated and real-world will need to be reconciled through deep transfer learning techniques (e.g. generative adversarial networks) to enable a smooth transition  ... 
doi:10.1007/s11263-018-1073-7 fatcat:go2kohdljzfmfemp5z536p2qhy
« Previous Showing results 1 — 15 out of 4,894 results