WORKSHOPS

Jan. 2027 - April 2027

• Coatings and surface engineering technologies
• Corrosion and corrosion protection
• Green chemistry
• Catalysis
• Safety in R/D institutions

WORKSHOPS

Mai 2027 - Aug. 2027

• Entrepreneurship
• Ammoniac economy
• Recycling technologies and waste valorisation
• High temperature Materials
• Optoelectrics

AI and Machine Learning for the Physical Sciences

1. Machine Learning for Scientists: Supervised, Unsupervised, and Probabilistic Learning (04.09.2026)
Foundations & Scientific Insight Expansion: Building core ML understanding tailored to scientific problems, enabling robust pattern discovery, uncertainty quantification, and data-driven hypothesis generation.

2. Neural Networks and Deep Learning: Architecture and Physical Intuition (11.09.2026)
Scalable Modeling & Physical Reasoning Integration: Designing deep architectures informed by physical laws to improve interpretability, efficiency, and generalization in complex scientific systems.

3. Scientific Data: Representations, Descriptors, and Preprocessing (18.09.2026)
Data Quality & Feature Engineering Optimization: Transforming raw experimental and simulation data into meaningful, ML-ready representations that enhance model accuracy and reproducibility.

4. Gaussian Processes and Bayesian Optimisation for Science (25.09.2026)
→ Uncertainty-Aware Optimization & Efficient Exploration: Leveraging probabilistic models to guide experiments, reduce sample costs, and accelerate discovery under uncertainty.

5. Graph Neural Networks for Molecules and Materials (02.10.2026)
Structure-Aware Learning & Atomic-Level Prediction: Utilizing graph-based representations to capture relational information in molecules and materials for accurate property prediction and design.

6. Machine Learning Interatomic Potentials and Force Fields (09.10.2026)
High-Fidelity Simulation Acceleration: Replacing expensive quantum calculations with ML-driven potentials to enable large-scale, near-ab initio accuracy simulations.

7. Physics-Informed Machine Learning (16.10.2026)
Hybrid Modeling & Constraint Integration: Embedding physical laws into ML models to improve data efficiency, extrapolation, and trustworthiness in scientific predictions.

8. Generative Models for Molecules and Materials (23.10.2026)

compounds and materials with desired properties using generative AI approaches such as VAEs, GANs, and diffusion models.

9. AI for Electrochemical Data Analysis (30.10.2026)
Signal Interpretation & Process Optimization: Applying AI to extract insights from complex electrochemical signals, improving battery diagnostics and reaction understanding.

10. AI-Accelerated Materials Discovery (06.11.2026)
End-to-End Discovery Pipelines: Integrating ML across simulation, experimentation, and screening to drastically shorten the materials innovation cycle.

11. AI for Spectroscopy and Characterisation (13.11.2026)
Automated Interpretation & High-Throughput Analysis: Using AI to decode complex spectra and imaging data, enabling faster and more accurate material characterization.

12. Large Language Models and AI Tools for Research Practice (20.11.2026)
Research Augmentation & Knowledge Automation: Leveraging LLMs for literature review, coding, documentation, and scientific communication to enhance productivity and insight generation.

13. Autonomous Experimentation and Self-Driving Laboratories (27.11.2026)
Closed-Loop Discovery & Robotic Automation: Combining AI, robotics, and real-time feedback to create fully automated labs capable of independent experimentation and optimization.

14. Explainability and Interpretability in Scientific ML (04.12.2026)
→ Transparent Models & Scientific Trust: Developing methods to understand ML decisions, ensuring alignment with physical principles and enabling actionable insights.

15. Reproducible ML in Science: Workflows, Benchmarks, and Best Practices (11.12.2026)
Robust Research & Standardized Pipelines: প্রতিষ্ঠuting reproducible workflows, data/version control, and benchmarking standards to ensure reliability and comparability of results.

16. Ethics, Bias, and Responsible AI in the Physical Sciences (18.12.2026)
Responsible Innovation & Risk Mitigation: Addressing bias, fairness, environmental impact, and ethical considerations in deploying AI within scientific and industrial contexts.