Skip to main content

Advanced Artificial Intelligence

About the Group

The research group conducts basic and applied research in all aspects of artificial intelligence. In particular, it contributes to general knowledge in the following sub-disciplines: machine learning, computer vision, image processing, computational intelligence, hyper-heuristics, data visualization, and applies them to problem solving within contexts such as health, business, public safety, computer security, among others.

Because of its intersection with these areas, the group also influences the development and application of data science.
 

Research lines

• Machine learning    
• Computational intelligence and hyper-heuristics  
• Data science and applied mathematics
• Biomedical engineering

Leader

Raúl Monroy Borja - raulm@tec.mx
 

Co-leader

Juana Julieta Noguez Monroy - jnoguez@tec.mx
 

Members

César Torres Huitzil
Edgar Covantes Osuna
Gilberto Ochoa Ruiz
Gildardo Sánchez Ante
Hugo Terashima Marín
Iván Mauricio Amaya Contreras
Jesús Guillermo Falcón Cardona
Jorge Mario Cruz Duarte
José Antonio Cantoral Ceballos
José Carlos Ortíz Bayliss
Luciano García Bañuelos
Luis Ángel Trejo Rodríguez
Miguel González Mendoza
Rajesh Roshan Biswal
Salvador Miguel Hinojosa Cervantes
Santiago Enrique Conant Pablos


Postdoctoral researchers

Mariano Vargas Santiago
Bárbara Cervantes González
Víctor Adrián Sosa Hernández
Octavio Loyola González
Joanna Alvarado Uribe
Nestor Velasco Bermeo
José Benito Camiña Prado
Ari Yair Barrera Animas
Iván Mauricio Amaya Contreras
Alejandro Rosales Pérez
Andrés Eduardo Gutiérrez Rodríguez
Frumencio Olivas Álvarez
Jorge Mario Cruz Duarte

Most relevant publications

• Cruz-Duarte, J. M., Amaya, I., Ortiz-Bayliss, J. C., Conant-Pablos, S. E., Terashima-Marín, H., & Shi, Y. Hyper-heuristics to customise metaheuristics for continuous optimisation. Swarm and Evolutionary Computation, 66. 2021.

• Diaz-Ramos, R.E.; Gomez-Cravioto, D.A.; Trejo, L.A.; Figueroa López, C.; Medina-Pérez, M.A. Towards a Resilience to Stress Index Based on Physiological Response: A Machine Learning Approach. Sensors, 21, 8293. Special Issue: Sensors and Digital Solutions for Human Health and Health Risk Monitoring. 2021.

• Hinojosa, S., Oliva, D., Cuevas, E. et al. Reducing overlapped pixels: a multi-objective color thresholding approach. Soft Comput 24, 6787–6807 (2020).

• Pérez-Torres, R., Torres-Huitzil, C. and Galeana-Zapién, H.: An On-Device Cognitive Dynamic Systems Inspired Sensing Framework for the IoT. IEEE Commun. Mag. 56(9): 154-161. 2018

• Ortiz-Bayliss, J.-C., Amaya, I. and Cruz-Duarte, J.-M., Gutierrez-Rodriguez, A.-E., Conant Pablos, S.-E. and Terashima-Marin, H. A General Framework based on Machine Learning for Algorithm Selection in Constraint Satisfaction Problems. Applied Sciences, 11(6), pp. 1-16. 2021.

• Martínez-Díaz, Y., Nicolás-Díaz, M., Méndez-Vázquez, H., Luevano, L. S., Chang, L., Gonzalez-Mendoza, M., Sucar, L. E. Benchmarking lightweight face architectures on specific face recognition scenarios. Artif Intell Rev 54, 6201–6244. 2021.

• Oliva, D., Hinojosa, S., Osuna-Enciso, V., Cuevas, E., Pérez-Cisneros, M., & Sanchez-Ante, G. Image segmentation by minimum cross entropy using evolutionary methods. Soft Computing, 23(2), 431-450. 2019.

• Falcón-Cardona, J.-G., Ishibuchi, H., Coello Coello, C.-A. and Michael Emmerich. Título del artículo: On the Effect of the Cooperation of Indicator-Based Multi-objective Evolutionary Algorithms. Revista: IEEE Transactions on Evolutionary Computation. 25(4): 681-695. 2021.

• Angeles-Ceron, J.-C., Ochoa-Ruiz, G., Chang, L. and Ali, S. Real-time Instance Segmentation of Surgical Instruments using Attention and Multi-scale Feature Fusion, Under review at Medical Image Analysis, 2021.

• J. Rodríguez, J.-I. Mata-Sánchez, R. Monroy, O. Loyola-González, A. López-Cuevas. A one-class classification approach for bot detection on Twitter. Computers and Security 91, April 2020, article 101715 © Elsevier, 2020 ** 2021 Rómulo Garza's best paper award **

 

Most relevant projects

Intelligence Artificielle pour la generation de microfictions littéraires (GenMicFic).
Raúl Monroy Borja (co-principal investigator). SEP - CONACYT - ANUIES - ECOS NORD Francia. Generación de microficciones, un género literario, mediante el uso de aprendizaje profundo, en particular transformadores (GPT-3) y SBERT. El mecanismo considera los elementos de un relato (inicio, desarrollo y conclusión), así como una variedad  de condiciones, tales como coherencia y narrativa en tercera persona.

Proyecto Context-Aware Video Detection and Interpretation of Suspicious Behavior Using Distributed Robust Deep Learning
In collaboration with Dr. Paul Rad and Dr. David Han of the University of Texas at San Antonio.
Principal Investigator: Hugo Terashima Marín. Fondo Tec-UTSA $40000USD.
The proposed intelligent image/video analytics technology will allow users to: Search for images/videos by identifying not only objects but also structured relationships and attributes involving these objects, like: “find an image or video in which a man is carrying a bomb in the airport”; Understand the consistency between textual features and visual contents; Recognize all of the connections and effects between deep extracted features and retrieve images by considering all the dependencies of features; Identify and tag meaningful high-level situation descriptions presented in the retrieved video, like a person threatening another with a weapon, a person playing with a pet, etc.; and Based on results of previous objectives, work along a methodology for searching for meaningful features in order to detect suspicious behavior in video for a particular context.

One Step closer to mental health: promptly detection of depression with wearables technology and voice analysis.
Principal Investigator: Luis Angel Trejo Rodríguez.
Symptoms of depression can be detected with machine learning algorithms techniques, using heart rate variability, sleeping patterns, type of personality and physical activity as attributes. The main objective is to develop an intelligent system to detect in time, through wearables, voice analysis, and machine learning, levels of depression of high risk in the final user.

Space-time laminar computing: event-based spike neuromorphic processors for sensory computation.
Principal Investigator: César Torres Huitzil.
Computing in its diverse forms has become essential to most aspects of modern life, but it still fails in some of the basic tasks that biological systems (humans) perform easily and efficiently, such as perception, motor control, language processing, etc. Far beyond “intelligence”, computing based on current single-processor architectures and the associated semiconductor technology are facing fundamental physical limits (scalability, power consumption, process variations, noise margins, and fault tolerance, etc.) that prevent them achieving better performances only through improved processor technologies. The aim of this project is to provide the knowledge to develop a new class of event-driven spike neuroprocessors aimed at power efficient sensory computation under a neuro-inspired space-time laminar computational framework.

Feature transformation for improving characterization of combinatorial optimization problems. Sectoral research fund for general education (CONACyT).
José Carlos Ortiz Bayliss, Research Associate.
The project seeks to improve the descriptive capabilities of hyper-heuristics by transforming the characterization of the problems they solve.

Robust Surgical Tool Segmentation, Tracking and Depth Perception.
Gilberto Ochoa Ruiz in collaboration with Dr. Sharib Ali, from the University of Leeds (United Kingdom).
To develop new datasets, schemes and models for implement robust and real-time computer vision methods for Computer Integrated Surgery (CIS) applications and procedural quality assessment purposes. Students: Mansoor Ali Teevno.

RECONDITE: Deep learning and image analysis methods for improving the endoscopic identification of kidney stones composition.
Gilberto Ochoa Ruiz in collaboration with Prof. Christian Daul the Centre de Recherche en Automatique de Nancy, CRAN (France) and the Institut National de la Santé et de la Recherche Medicale (INSERM).
To investigate deep learning algorithms for automatically classifying in vivo kidney stones from endoscopy images. Students: Francisco Lopez Tiro, Daniel Florez Araiza.

ISOLATE: SegmentatIon and claSsification Of vascuLar pATtern symmEtries on cerebral vessels using DL.
Gilberto Ochoa Ruiz in collaboration with Dr. Christian Mata Miquel and Prof. Enrique Benitez from the Biomedical Engineering Research Center (CREB, Barcelona) of the Universitat Politecnica de Catalunya (Spain) and the Hospital Sant Joan de Deu (Barcelona).
To develop novel CADx tools for aiding physicians in the diagnosis of CP. Various algorithms for vessel segmentation and skeletonization have been explored and tested. The results of these preprocessing methods are to be used for classifying vascular pattern asymmetries.

Business relationship

Arca Continental, Santiago E. Conant Pablos (investigador principal), IDEA: Innovación Mediante Ciencia de Datos e Inteligencia Artificial para Mejorar los Indicadores Clave de Negocio en Clientes del Canal Tradicional, 2019 - 2020.
Google Inc. - APRU, Raúl Monroy Borja (experto) AI for all, 2017 - 2018.
Google Inc., Raúl Monroy Borja (investigador principal) Formal Verification of Web Applications, 2015 - 2016.
NIC - México, Raúl Monroy Borja (investigador principal) Dynamic Networks and Metrics for Ad Efficiency Ratings, 2017 - 2019.
NIC - México, Raúl Monroy Borja (investigador principal) Countermeasures for DDoS Attacks Targeting the Domain Name System, 2017 - 2019.