Artificial Intelligence: Learn, Infer, Decide!


Artificial Intelligence (AI) refers to systems that display intelligent behaviour by analysing their environment and taking actions to achieve specific goals.


Artificial Intelligence (AI) refers to systems that display intelligent behaviour by analysing their environment and taking actions to achieve specific goals. A recent success story within AI comes from statistical learning where data is automatically processed to learn key relationships.

The subject consists in a creative combination of mathematics and programming, where the use of data is always at the centre. Generic computer programs are used, which are adapted to application specific circumstances by automatically adjusting parameters of the program based on observed so-called training data.

  • Approximate inference (ApI): many learning problems involve intractable integrals, requiring good approximation methods; includes Monte Carlo and variational inference.
  • Causal learning (CL): causality is a fundamental notion in science and engineering; causal learning establishes cause-effect relationships from observations that can be empirically tested for their accuracy.
  • Constraint programming (CP) is an AI approach to Optimisation: modelling languages, high-level constraints, high-level types for decision variables, symmetry breaking.
  • Explainable Artificial Intelligence (XAI): artificial intelligence models that follow the principles of transparency, interpretability, and explainability, and provide solutions that can be understood by humans.
  • Knowledge representation & reasoning (KR): automated reasoning and theorem proving (ATP), knowledge compilation (KC), logic programming (LP).
  • Large-scale optimisation in learning (LO): optimisation problems with a huge number of unknowns, common in e.g. deep learning.
  • Local search (LS): meta-heuristics, modelling languages, search languages, solver design, autonomous search.
  • Machine learning (ML) is about learning, reasoning, and acting based on data; in this way, machine learning is about programming by examples; important topics include clustering, deep learning, and inductive logic programming (ILP) for relational learning.
  • Pattern recognition (PR) is the automated recognition of patterns and regularities in data, typically by the use of ML.
  • Propositional satisfiability (SAT) and SAT modulo theories (SMT): trustworthy and verified solvers, proofs and certificates, competitions and evaluations.
  • Probabilistic modelling (PM): mathematical models capable of representing uncertainties.
  • Reinforcement learning (RL) studies algorithms capable of perceiving their environment, interpret it, take actions, and learn through trial and error.
  • Social robotics (SR): human-robot interaction, socially intelligent robots, social artificial intelligence, multimodal interaction.

  • 1DL010: Artificial Intelligence (7.5 credits): KR, ML, search
  • 1DL340: Artificial Intelligence (5 credits): KR, ML, search
  • 1DL360: Data Mining I (5 credits): ML, search
  • 1DL370: Data Mining (7.5 credits): ML, search
  • 1DL442: Combinatorial Optimisation and Constraint Programming (10 credits): CP, LS
  • 1DL451: Modelling for Combinatorial Optimisation (5 credits): CP, LS, SAT, SMT
  • 1DL481: Algorithms and Data Structures III 85 credits): LS, SAT, SMT
  • 1MD032: Intelligent Interactive Systems (5 credits): SR
  • 1MD039: Intelligent Interactive Systems (7.5 credits): SR
  • 1MD120: Deep Learning for Image Analysis (7.5 credits)
  • 1MD300: Social robotics and human-robot interaction (7.5 credits): SR
  • 1RT003: Advanced Probabilistic Machine Learning (7.5 credits)
  • 1RT700: Statistical Machine Learning (5 credits)