Embedded Files
Together with the ISPAMM Lab team, I collaborate with industry to connect academic research in artificial intelligence, machine learning, and signal processing with real-world technological challenges. These activities span different application domains and data modalities, combining methodological research with the development and evaluation of AI solutions for complex industrial problems.
Industrial collaboration also creates a valuable exchange between research and practice, enabling emerging methodologies to be explored in realistic scenarios while real-world problems, data, and constraints open new directions for scientific research.
We welcome opportunities to explore research collaborations and knowledge-exchange activities with companies interested in addressing complex technological challenges through advanced AI methodologies.
Areas of Collaboration
AI Modeling & Prediction
AI Modeling & Prediction
Machine learning methods for prediction, forecasting, classification, regression, retrieval, detection, and recognition across heterogeneous data and application scenarios.
Generative & Multimodal AI
Generative & Multimodal AI
Generative models for data synthesis, augmentation, reconstruction, transformation, and multimodal content generation, including methods for integrating and aligning information across different modalities.
Intelligent Systems
Intelligent Systems
AI methods for monitoring complex systems, identifying anomalous behaviors and operating conditions, and supporting data-driven analysis and decision-making in real-world environments.
Multisensor & Physical Data
Multisensor & Physical Data
Learning from heterogeneous information sources, including audio, images, video, time series, sensor measurements, spatial information, and data generated by complex physical systems.
The methodological nature of these activities allows the same AI principles to be investigated across different industrial domains, rather than being restricted to a specific application sector.
Emerging Directions
My current interests for future industrial collaborations extend toward AI systems capable of modeling, reasoning about, and interacting with complex physical environments.
Digital Twins: Data-driven and AI-enhanced representations of physical systems for monitoring, prediction, simulation, and decision support.
World Models: Learning representations of environments and their dynamics to support prediction, simulation, and intelligent interaction.
Physics-Informed AI: Integrating data-driven learning with physical knowledge, models, and constraints.
Embodied AI: Intelligent systems that perceive, reason about, and interact with physical environments.
Agentic AI: AI systems capable of planning, reasoning, and coordinating actions toward complex objectives.
Knowledge Exchange & AI Training
We also collaborate with companies through tailored seminars and short training activities on artificial intelligence and machine learning. Content can be adapted to specific technological interests, application domains, and organizational needs, combining methodological foundations with relevant real-world examples and use cases. Topics can range from established machine learning methodologies to emerging areas of AI, depending on the interests and technical background of the audience.
Report abuse