European Laboratory for Learning and Intelligent Systems Institute Campuses

At the Kumpula Campus of the European Laboratory for Learning and Intelligent Systems Institute, courses focus on theoretical foundations, probabilistic modeling, and scientific applications of AI. The offerings promote research excellence.

• Probabilistic Machine Learning: Bayesian networks, Gaussian processes, MCMC sampling, and uncertainty modeling.
• Advanced Deep Learning: Theory of generalization, adversarial robustness, and self-supervised learning.
• Statistical Learning Theory: VC dimension, PAC learning, empirical risk minimization, and high-dimensional statistics.
• Graphical Models: Directed and undirected models, inference algorithms, and structure learning.
• Causal Inference in AI: Counterfactuals, do-calculus, instrumental variables, and causal discovery.
• Natural Language Understanding: Compositional semantics, reasoning over text, and knowledge graphs.
• Reinforcement Learning Theory: Regret bounds, exploration-exploitation, and bandit algorithms.
• AI for Scientific Discovery: ML in physics, biology, climate modeling, and hypothesis generation.
• Interpretability and Fairness: Causal explanations, counterfactual fairness, and algorithmic auditing.
• Quantum Machine Learning: Quantum circuits, variational quantum algorithms, and hybrid models.

The campus features theoretical seminars and reading groups with ELLIS fellows. Students engage in proof-based projects and publish at top conferences like NeurIPS and ICML. Ties to natural sciences enable cross-disciplinary research in bioinformatics and neuroscience. Doctoral tracks delve into frontier topics like AI safety and trustworthy systems. Graduates pursue academia or research labs worldwide.

At the Otaniemi Campus of the European Laboratory for Learning and Intelligent Systems Institute, courses emphasize applied AI and engineering integration with intelligent systems. The program offers rigorous training in cutting-edge technologies.

• Machine Learning: Supervised and unsupervised learning, ensemble methods, feature engineering, and real-world applications in industry.
• Deep Learning Architectures: CNNs, RNNs, transformers, generative models, and optimization for large-scale deployment.
• Reinforcement Learning: Policy optimization, deep RL, multi-agent systems, and robotics applications.
• Computer Vision: Image recognition, 3D reconstruction, autonomous systems, and embedded vision.
• Natural Language Processing: Sequence modeling, large language models, multilingual AI, and conversational agents.
• Scalable AI Systems: Distributed computing, cloud ML, edge AI, and hardware-software co-design.
• AI for Robotics: Perception, planning, control, and human-robot interaction using ML.
• Optimization and Control: Convex optimization, stochastic gradient methods, and feedback systems.
• AI Ethics and Engineering: Responsible AI design, safety in autonomous systems, and regulatory compliance.
• Big Data Analytics: Spark, Hadoop integration with ML pipelines, and streaming data processing.

Hands-on labs utilize GPU clusters and simulation environments. Students collaborate on projects with Finnish tech firms and international partners. The curriculum supports interdisciplinary work with electrical engineering and design, fostering innovation in smart cities, autonomous vehicles, and sustainable tech. Advanced seminars cover neuromorphic computing and AI accelerators. This equips graduates for roles in R&D at companies like Nokia and Kone.