1.

Record Nr.

UNINA9910760269403321

Autore

Calvo Hiram

Titolo

Advances in Computational Intelligence : 22nd Mexican International Conference on Artificial Intelligence, MICAI 2023, Yucatán, Mexico, November 13–18, 2023, Proceedings, Part I / / edited by Hiram Calvo, Lourdes Martínez-Villaseñor, Hiram Ponce

Pubbl/distr/stampa

Cham : , : Springer Nature Switzerland : , : Imprint : Springer, , 2024

ISBN

3-031-47765-0

Edizione

[1st ed. 2024.]

Descrizione fisica

1 online resource (363 pages)

Collana

Lecture Notes in Artificial Intelligence, , 2945-9141 ; ; 14391

Altri autori (Persone)

Martínez-VillaseñorLourdes

PonceHiram

Disciplina

006.3

Soggetti

Artificial intelligence

Computers

Database management

Application software

Artificial Intelligence

Computing Milieux

Database Management System

Computer and Information Systems Applications

Lingua di pubblicazione

Inglese

Formato

Materiale a stampa

Livello bibliografico

Monografia

Nota di contenuto

Intro -- Preface -- Organization -- Contents - Part I -- Contents - Part II -- Machine Learning -- Stock Market Performance Analytics Using XGBoost -- 1 Introduction -- 2 Related Work -- 3 Methodology and Development -- 3.1 Dataset Description -- 3.2 Data Preparation -- 3.3 Using XGBoost for Stock Trend and Prices Prediction -- 3.4 Moving Average -- 4 Results and Discussion -- 4.1 Correlation of Stock -- 4.2 XGBoost Regressor -- 4.3 MACD -- 5 Conclusion -- References -- 1D Quantum Convolutional Neural Network for Time Series Forecasting and Classification -- 1 Introduction -- 2 Preliminaries -- 2.1 Quantum Computation -- 2.2 Quantum Circuits -- 2.3 Variational Quantum Circuits -- 3 1D Quantum Convolution -- 4 Results and Discussion -- 4.1 Time Series Forecasting -- 4.2 Time Series Classification Using PTB Dataset -- 5 Conclusions -- References -- Hand Gesture Recognition



Applied to the Interaction with Video Games -- 1 Introduction -- 2 Methodology -- 2.1 Hand Gesture Recognition -- 2.2 Video Game Application -- 3 Results -- 4 Conclusions and Future Work -- References -- Multiresolution Controller Based on Window Function Networks for a Quanser Helicopter -- 1 Introduction -- 2 Application of the Control Scheme to the Helicopter Model -- 2.1 Dynamic Identification -- 2.2 Proportional Multi-resolution Controller -- 2.3 Autotune of the Gains -- 3 Results -- 3.1 Open-Loop Simulation Results: Identification Process -- 3.2 Closed-Loop Simulation Results -- 3.3 Comparative Between PMR and PID Controllers -- 4 Conclusions and Future Work -- References -- Semi-supervised Learning of Non-stationary Acoustic Signals Using Time-Frequency Energy Maps -- 1 Introduction -- 2 Methods -- 2.1 STFT Maps -- 2.2 Dimensionality Reduction Using PCA -- 2.3 Background Subtraction -- 3 Description of Proposed Method -- 4 Acoustic Non-stationary Signals.

4.1 Signals Acquisition -- 5 Classifier Results -- 5.1 Training and Classification -- 6 Conclusion -- References -- Predict Email Success Based on Text Content -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Description of the Dataset -- 3.2 Building and Training the Model -- 3.3 Model Evaluation -- 4 Results and Discussion -- 5 Conclusions -- References -- Neural Drone Racer Mentored by Classical Controllers -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Neural Pilot -- 3.2 Proportional-Integral Controller -- 3.3 Model Predictive Controller -- 3.4 Active Disturbance Rejection Control -- 3.5 Training Process -- 4 Experimental Framework -- 5 Conclusions -- References -- Eye Control and Motion with Deep Reinforcement Learning: In Virtual and Physical Environments -- 1 Introduction -- 2 Proposed Solution -- 2.1 Deep Reinforcement Learning and the NN's Structure -- 2.2 Optimising the Learning Process -- 3 The Experiment -- 3.1 Optimizing Training with Image Detection -- 3.2 Testing with a Real Camera -- 3.3 Performance Metrics -- 4 Training Results and Discussion -- 4.1 Testing with a Real Camera -- 5 Conclusions -- References -- Fingerspelling Recognition in Mexican Sign Language (LSM) Using Machine Learning -- 1 Introduction -- 2 Related Work -- 3 Methods -- 3.1 Data Capturing -- 3.2 Keypoints Detection -- 3.3 Keypoints Processing -- 3.4 Classification -- 4 Results -- 5 Conclusions -- References -- Load Demand Forecasting Using a Long-Short Term Memory Neural Network -- 1 Introduction -- 1.1 Artificial Neural Networks -- 2 Methodology -- 2.1 Data Provided by CENACE -- 2.2 LSTM Neural Network -- 2.3 Processing Data -- 2.4 Double-Input Long Short-Term Memory Neural Network -- 2.5 Triple-Input Long Short-Term Memory Neural Network -- 2.6 4-input Long Short-Term Memory Neural Network -- 2.7 Forecasting Errors -- 3 Results.

3.1 Double-Input LSTM -- 3.2 Triple-Input LSTM -- 3.3 Four-Input LSTM -- 4 Discussion and Conclusion -- References -- Computer Vision and Image Processing -- Benchmark Analysis for Backbone Optimization in a Facial Reconstruction Model -- 1 Introduction -- 2 Related Work -- 2.1 Face Reconstruction Models -- 2.2 Lightweight Backbone Architectures for Facial Reconstruction -- 3 Methodology -- 4 Results and Discussion -- 5 Conclusion -- References -- T(G)V-NeRF: A Strong Baseline in Regularized Neural Radiance Fields with Few Training Views -- 1 Introduction -- 2 Related Work -- 2.1 Neural Radiance Fields -- 2.2 Training a NeRF with (very) Few Images -- 3 Proposed Regularization Framework -- 3.1 Depth Map Regularization -- 3.2 Second-Order Regularization -- 3.3 Occlusion Regularization -- 4 Experimental Results -- 4.1 Implementation Details -- 4.2 Ablation Results -- 4.3 Quantitative Results -- 4.4 Qualitative Results -- 5



Conclusions -- References -- Nonlinear DIP-DiracVTV Model for Color Image Restoration -- 1 Introduction -- 2 DIP-Reg Models for Color Image Restoration -- 2.1 DIP-Reg Models -- 2.2 Euclidean Models: DIP-TV and DIP-VTV -- 2.3 Riemannian Model: DIP-RiemannVTV -- 3 Dirac Vectorial Total Variations of a Color Image -- 3.1 The Euclidean Dirac Operator and Its Main Properties -- 3.2 Dirac Operators for Color Images -- 4 Experiments -- 4.1 Numerical Scheme -- 4.2 Application to Denoising -- 4.3 Application to Deblurring -- 5 Conclusion -- References -- An Efficient Facial Verification System for Surveillance that Automatically Selects a Lightweight CNN Method and Utilizes Super-Resolution Images -- 1 Introduction -- 2 Related Work -- 3 Super-Resolution Methods -- 4 Light Facial Verification Methods -- 5 Proposed Dynamic Scaling with Super-Resolution Methods -- 6 Proposed Lightweight Facial Verification System -- 7 Experiments.

7.1 Implementation Details -- 7.2 Datasets -- 7.3 Selection of Super-Resolution Method -- 7.4 Evaluation with the Custom Dataset -- 7.5 Evaluation of the LFVS with the Datasets -- 8 Discussion -- 9 Conclusion -- References -- Nonlinear L2-DiracVTV Model for Color Image Restoration -- 1 Introduction -- 2 The L2-VTV Model for Color Image Restoration -- 2.1 Variational Models for Image Restoration -- 2.2 The L2-VTV Model and Its Solutions -- 2.3 Primal-Dual Algorithm -- 3 Non Linear L2-DiracVTV Model for Color Image Restoration -- 3.1 Dirac Operators for Color Images -- 3.2 Weighted Dirac Operators for Color Images -- 3.3 Linear Weighted Dirac Vectorial Total Variation and Its Dual Formulation -- 3.4 Variational Models for Color Image Restoration -- 4 Experiments -- 4.1 On the Choice of the Color Space -- 4.2 Denoising -- 4.3 Deblurring -- 5 Conclusion -- References -- An FPGA Smart Camera Implementation of Segmentation Models for Drone Wildfire Imagery -- 1 Introduction -- 2 State-of-the-Art -- 2.1 Segmentation Models for Wildfire Detection and Characterization -- 2.2 Smart Camera Implementations for Computer Vision -- 3 Proposed Method -- 3.1 General Overview of the Optimization Approach -- 3.2 Dataset: Corsican Fire Database -- 3.3 Segmentation Model Training -- 3.4 Optimization -- 3.5 Proposed FPGA-Based Smart Camera System -- 4 Results and Discussion -- 4.1 Comparison Metrics -- 4.2 Quantitative Results -- 4.3 Qualitative Results -- 5 Conclusions -- References -- Intelligent Systems -- An Argumentation-Based Approach for Generating Explanations in Activity Reasoning -- 1 Introduction -- 2 Background -- 3 Proposal -- 3.1 Human Activity Framework and Local Selection -- 3.2 Global Selection -- 4 Generating Explanations -- 5 Theoretical Evaluation -- 6 Conclusions and Future Work -- References.

A Decision Tree Induction Algorithm for Efficient Rule Evaluation Using Shannon's Expansion -- 1 Introduction -- 2 Background -- 3 Proposal -- 4 Experimental Results -- 5 Conclusions -- References -- Reasoning in DL-LiteR Based Knowledge Base Under Category Semantics -- 1 Introduction -- 2 Set-Theoretical Semantics of DL-LiteR -- 3 Category-Theoretical Semantics of DL-LiteR -- 4 Category-Theoretical Satisfiability for DL-LiteR -- 5 Conclusion -- References -- Applying Genetic Algorithms to Validate a Conjecture in Graph Theory: The Minimum Dominating Set Problem -- 1 Introduction -- 2 Background -- 3 The Conjecture to Be Verified -- 3.1 Generalized Quadrangle -- 3.2 The Minimum Dominating Set Problem -- 3.3 The Conjecture -- 4 Rank Genetic Algorithm for Finding Minimal Dominating Set -- 4.1 Solution Representation -- 4.2 Fitness Function -- 4.3 The Rank GA Operators -- 5 Results -- 6 Conclusions -- References -- Random Number Generator Based on Hopfield Neural Network with Xorshift and Genetic Algorithms -- 1 Introduction -- 2



NIST Testing Module -- 3 Related Works -- 3.1 Approaches Based on Chaotic Systems -- 3.2 Hybrid Pseudo Random Number Generator (PRNG) -- 3.3 Genetic Algorithms -- 4 PRNG Based on Hopfield Neural Network and Xorshift -- 5 PRNG Based on Genetic Algorithms -- 5.1 Selection -- 5.2 Loop -- 6 Tests and Results: PRNG Based on Hopfield Neural Network and Xorshift -- 6.1 Results of the NIST Module -- 6.2 Diehard Test Suite Results -- 7 Tests and Results: PRNG Based on Genetic Algorithms -- 7.1 Models to Compare with Ours -- 7.2 DieHarder Suite of Random Number Generator Tests -- 8 Discussion -- 9 Conclusion -- References -- Using Compiler Errors Messages to Feedback High School Students Through Machine Learning Methods -- 1 Introduction -- 2 Methodology -- 3 Classifier Development -- 3.1 Corpus Generation -- 3.2 Data Cleaning.

3.3 Vectorization.

Sommario/riassunto

The two-volume set LNAI 14391 and 14392 constitutes the proceedings of the 22nd Mexican International Conference on Artificial Intelligence, MICAI 2023, held in Yucatán, Mexico, in November 2023. The total of 49 papers presented in these two volumes was carefully reviewed and selected from 115 submissions. The proceedings of MICAI 2023 are published in two volumes. The first volume, Advances in Computational Intelligence, contains 24 papers structured into three sections: – Machine Learning – Computer Vision and Image Processing – Intelligent Systems The second volume, Advances in Soft Computing, contains 25 papers structured into three sections: – Natural Language Processing – Bioinformatics and Medical Applications – Robotics and Applications.