559 Hits in 9.4 sec

Learning Model Checking and the Kernel Trick for Signal Temporal Logic on Stochastic Processes [article]

Luca Bortolussi, Giuseppe Maria Gallo, Jan Křetínský, Laura Nenzi
2022 arXiv   pre-print
We demonstrate this consequence and its advantages on the task of predicting (quantitative) satisfaction of STL formulae on stochastic processes: Using our kernel and the kernel trick, we learn (i) computationally  ...  We introduce a similarity function on formulae of signal temporal logic (STL).  ...  Having discussed the possibility of learning model checking, the main potential of our kernel (and generally introducing kernels for any further temporal logics) is that it opens the door to efficient  ... 
arXiv:2201.09928v1 fatcat:ivs5r4jvxzgxdm5wahxpybmn5i

AI in Finance: Challenges, Techniques and Opportunities [article]

Longbing Cao
2021 arXiv   pre-print
We then structure and illustrate the data-driven analytics and learning of financial businesses and data.  ...  The comparison, criticism and discussion of classic vs. modern AI techniques for finance are followed.  ...  Kernel tricks for complex problem modeling; scalable; metric properties; etc. Kernel selection; calibration; training; interpretability; and actionability; etc.  ... 
arXiv:2107.09051v1 fatcat:g62cz4dqt5dcrbckn4lbveat3u

Event detection and recognition for semantic annotation of video

Lamberto Ballan, Marco Bertini, Alberto Del Bimbo, Lorenzo Seidenari, Giuseppe Serra
2010 Multimedia tools and applications  
Research on methods for detection and recognition of events and actions in videos is receiving an increasing attention from the scientific community, because of its relevance for many applications, from  ...  Event detection and recognition requires to consider the temporal aspect of video, either at the low-level with appropriate features, or at a higher-level with models and classifiers than can represent  ...  Acknowledgement This work is partially supported by the EU IST IM3I Project (Contract FP7-222267).  ... 
doi:10.1007/s11042-010-0643-7 fatcat:7x6x4r3n6rhnnofpqhbelgzc6y

Towards Comprehensive Foundations of Computational Intelligence [chapter]

Włodzisław Duch
2007 Studies in Computational Intelligence  
The need to understand data structures leads to techniques for logical and prototype-based rule extraction, and to generation of multiple alternative models, while the need to increase predictive power  ...  Several proposals for CI foundations are discussed: computing and cognition as compression, meta-learning as search in the space of data models, (dis)similarity based methods providing a framework for  ...  One popular way of creating highly-dimensional representations without increasing computational costs is by using the kernel trick [155] .  ... 
doi:10.1007/978-3-540-71984-7_11 fatcat:vlfuzhxmgbbqlptp4wnpdty4ae

ConvLSTM Coupled Economics Indicators Quantitative Trading Decision Model

Yong Qi, Hefeifei Jiang, Shaoxuan Li, Junyu Cao
2022 Symmetry  
spatial and temporal features within the data.  ...  The 2016–2021 Bitcoin value dataset published on Kaggle was used for simulated investment.  ...  an autoregressive model with a stochastic process.  ... 
doi:10.3390/sym14091896 fatcat:pvxdjeof5rb7bdqjpvfgtbe7ui

Scalable high-resolution forecasting of sparse spatiotemporal events with kernel methods: a winning solution to the NIJ "Real-Time Crime Forecasting Challenge" [article]

Seth Flaxman and Michael Chirico and Pau Pereira and Charles Loeffler
2019 arXiv   pre-print
Model hyperparameters including quality of RKHS approximation, spatial and temporal kernel lengthscales, number of autoregressive lags, bandwidths for smoothing kernels, as well as cell shape, size, and  ...  While the smoothing kernels capture the two main approaches in current use in the field of crime forecasting, kernel density estimation (KDE) and self-exciting point process (SEPP) models, the RKHS component  ...  Special thanks to our systems administrators: Tony Vo (University of Pennsylvania) and Stuart McRobert (Oxford).  ... 
arXiv:1801.02858v4 fatcat:mwbssmnrlzhgdedsmp6k5gxony

Specification-Based Monitoring of Cyber-Physical Systems: A Survey on Theory, Tools and Applications [chapter]

Ezio Bartocci, Jyotirmoy Deshmukh, Alexandre Donzé, Georgios Fainekos, Oded Maler, Dejan Ničković, Sriram Sankaranarayanan
2018 Lecture Notes in Computer Science  
In the context of discrete systems, software or digital hardware, formalisms such as temporal logic or regular expressions are commonly used.  ...  The activity of simulating a system and checking its behaviour is part of the verification and validation process whose goal is to ensure, as much as possible, that the system behaves as expected and to  ...  Ničković acknowledge the partial support of the EU ICT COST Action IC1402 on Runtime Verification beyond Monitoring (ARVI) and of the HARMONIA (845631) project, funded by a national Austrian grant from  ... 
doi:10.1007/978-3-319-75632-5_5 fatcat:2m52qvfax5cn3fgyhcl3piy2ui

Recent Advances in Anomaly Detection Methods Applied to Aviation

Luis Basora, Xavier Olive, Thomas Dubot
2019 Aerospace (Basel)  
After a brief introduction to the main traditional data-driven methods for anomaly detection, we review the recent advances in the area of neural networks, deep learning and temporal-logic based learning  ...  flight trajectories and sensor data is sequential, or temporal.  ...  and deep learning as well as temporal logic based learning.  ... 
doi:10.3390/aerospace6110117 fatcat:kprkb643xrhcnmjy2c2lbzoa7m

A Survey on Machine Learning for Optical Communication [Machine Learning View] [article]

M. A. Amirabadi
2019 arXiv   pre-print
Machine Learning (ML) for Optical Communication (OC) is certainly a hot topic emerged recently and will continue to raise interest at least for the next few years.  ...  This view could be really helpful because only OC experts work on ML for OC, and they are not ML experts, so it could really help them to have a comprehensive view on the ML subjects implantable in OC.  ...  calculate distance, and c. finding neighbors and voting for labels. (based on some logical rules).  ... 
arXiv:1909.05148v1 fatcat:t635duaufrhohb4mteophpjqra

On the Adversarial Robustness ofGaussian Processes

Andrea Patane
2020 Zenodo  
Employing the central limit theorem for stochastic processes, we then demonstrate how the derived bounds can also be used for the adversarial analysis of infinitely-wide deep BNN architectures.  ...  On the other hand, adversarial robustness is concerned with local stability of the model decision, and is strictly correlated with bounds on the predictive posterior distribution of the model.  ...  They tackle the problem of synthesising a safe controller for the GP model, and identify a convex subset of the probabilistic signal temporal logic, which allows them to solve the problem formally by using  ... 
doi:10.5281/zenodo.5092158 fatcat:cullpr4i7zf4ljenafmhbmvu6m

Tunnel ventilation control via an actor-critic algorithm employing nonparametric policy gradients

Baeksuk Chu, Daehie Hong, Jooyoung Park
2009 Journal of Mechanical Science and Technology  
In this research, a reinforcement learning (RL) method based on the actor-critic architecture and nonparametric policy gradients is applied as the control algorithm.  ...  The appropriate operation of a tunnel ventilation system provides drivers passing through the tunnel with comfortable and safe driving conditions.  ...  One of popular control methods for such systems is fuzzy logic control, and there have been many studies for tunnel ventilation control using fuzzy logic [3] [4] [5] .  ... 
doi:10.1007/s12206-008-0924-5 fatcat:ysxuf7ly2zd7rkq44f7kqfv4au

Generative Adversarial Networks-Based Semi-Supervised Automatic Modulation Recognition for Cognitive Radio Networks

Mingxuan Li, Ou Li, Guangyi Liu, Ce Zhang
2018 Sensors  
These two technical improvements effectively avoid nonconvergence and mode collapse problems caused by the complexity of the radio signals.  ...  Here, a semi-supervised learning method based on adversarial training is proposed which is called signal classifier generative adversarial network.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s18113913 fatcat:45glnftoqfdfnfaecsk6vyu2oe

Graph Deep Learning: State of the Art and Challenges

S. Georgousis, M. P. Kenning, X. Xie
2021 IEEE Access  
We identify four major challenges in graph deep learning: dynamic and evolving graphs, learning with edge signals and information, graph estimation, and the generalization of graph models.  ...  The last half-decade has seen a surge in deep learning research on irregular domains and efforts to extend convolutional neural networks (CNNs) to work on irregularly structured data.  ...  modeling the temporal dynamics of the signal and for prediction [12] , [13] , [18] , [98] .  ... 
doi:10.1109/access.2021.3055280 fatcat:7ruskzkdkjgkfkia7drmww6lse

A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting

Erick C. Bezerra, Pierre Pinson, Ruth P. S. Leao, Arthur P. S. Braga
2021 IEEE Access  
The proposed method is based on a competitive tracking method, and the algorithm deals with some common difficulties of kernel methods, e.g., the increasing kernel matrix size associated with time and  ...  The results of the proposed method are compared with those provided by other kernel-based versions of the recursive least-squares algorithm, an online version of the extreme learning machine method, and  ...  temporal information based on computational fluid dynamic (CFD) models.  ... 
doi:10.1109/access.2021.3099999 fatcat:2ey4dcdumnew7a3etyd2kvvlce

Beyond ImageNet: Deep Learning in Industrial Practice [chapter]

Thilo Stadelmann, Vasily Tolkachev, Beate Sick, Jan Stampfli, Oliver Dürr
2019 Applied Data Science  
While deep neural networks have become popular primarily for image classification tasks, they can also be successfully applied to other areas and problems with some local structure in the data.  ...  We will focus on convolutional neural networks (CNNs), which have since the seminal work of Krizhevsky et al. (2012) revolutionized image classification and even started surpassing human performance on  ...  Acknowledgements The authors are grateful for the support by CTI grants 17719.1 PFES-ES, 17729.1 PFES-ES and 19139.1 PFES-ES.  ... 
doi:10.1007/978-3-030-11821-1_12 fatcat:5uo3qf2uw5e2dlzkjvtapynbhu
« Previous Showing results 1 — 15 out of 559 results