Coursework

This page gives an overview of the MSc Logic coursework I have completed or planned at the University of Amsterdam, grouped by research area rather than by semester. I am currently in my second year in period 7 of the programme.

My main interests are:

Quantum computing & quantum information

NoteHow this fits my interests

This group collects courses that underpin my main focus on quantum computing and quantum information, including both algorithmic / information-theoretic courses and physics-oriented courses on hardware and many-body methods.

  • Near-Term Quantum Computing – Kareljan Schoutens (period 2, completed): Hands-on work with NISQ devices and simulators using Qiskit, focusing on algorithms such as VQE and QAOA, noise models, and error-mitigation strategies. Course project: VQA for the vibrational spectrum of SO₂.
  • Full-Stack Quantum Computing – John van de Wetering (period 4, completed): The “full stack” from abstract quantum circuits down to fault-tolerant implementations, with an emphasis on ZX-calculus, stabiliser formalism, compilation, and quantum error correction.
  • Quantum Hardware – Arghavan Safavi-Naini (period 4, completed): Physical realisations of qubits and interacting two-level systems using ions and neutral atoms; decoherence and noise processes, and an overview of leading platforms for quantum computation and sensing.
  • Advanced Numerical Methods in Many Body Physics – Philippe Corboz (period 5, completed): Computational methods for classical and quantum many-body systems, including Monte Carlo algorithms and tensor-network techniques such as DMRG.
  • Quantum Cryptography – Florian Speelman (period 5, completed): Security notions and protocols in quantum cryptography (e.g. QKD, oblivious transfer, bit commitment, secret sharing) and quantum-hard assumptions for post-quantum cryptography.
  • Quantum Theory of Molecules and Matter – Wybren Buma (period 7, in progress): Applications of quantum mechanics to atoms, molecules, solids, and spectroscopy, including angular momentum, symmetry and group theory, approximation methods, atomic structure and spectra, molecular orbital theory, and band theory.
  • Quantum Computing for Chemistry and Physics – Freek Witteveen (period 8, planned): Quantum algorithms for the simulation of quantum physics and chemistry on fault-tolerant quantum computers, including many-body system modelling, qubit mappings and Hamiltonian encodings, Hamiltonian simulation techniques, and algorithms for preparing and analysing ground and thermal states.

Artificial intelligence, machine learning & interpretability

NoteHow this fits my interests

These courses and projects support my secondary focus on AI and mechanistic interpretability, connecting general-purpose machine learning and deep learning with language models and model explainability.

  • Machine Learning and Language Models – Martha Lewis (period 1, completed): Supervised and unsupervised learning, basic reinforcement learning, and applications to language modelling; implementation of key methods in Python and standard ML libraries.
  • Deep Learning 1 – Pascal Mettes (period 2, completed): Core deep-learning architectures and training methods, including CNNs, transformers, graph neural networks, generative models, and self-supervised learning.
  • Mechanistic interpretability of variable assignment in a Transformer-based model – Fausto Carcassi (period 3, completed): Independent group research project on mechanistic interpretability of a Transformer language model using causal interventions on the residual stream to study how variable assignment in Python code is represented. More details on the projects page.
  • Reinforcement Learning – Herke van Hoof (period 4, completed): Value-based and policy-based methods, approximate and deep RL for discrete and continuous control problems, and critical evaluation of RL experiments.
  • Interpretability & Explainability in AI – Willem Zuidema (period 6, completed): Post-hoc interpretability tools (e.g. saliency, attribution, probing, influence functions), explainable-by-design models, constrained deep learning, and evaluation of interpretability techniques.
  • Knowledge Representation and Reasoning – Ronald de Haan (period 7, in progress): Logical approaches to knowledge representation, reasoning, problem solving, and search in AI, including reasoning algorithms such as DPLL, answer set programming, and the expressivity and computational complexity of reasoning.

Computer science & logic

NoteHow this fits my interests

These courses provide theoretical computer science and logic foundations that support my work in quantum computing and AI.

  • Logic, Language and Computation – Nick Bezhanishvili (period 1–2, completed): Survey of research lines in logic, language, and information within the ILLC, including guest lectures and individual meetings with researchers and PhD students.
  • Information Theory – Nicolas Resch (period 2, completed): Shannon entropy and mutual information, data compression and channel coding theorems, zero-error information theory, and information-theoretic security.
  • Computational Complexity – Ronald de Haan (period 4, completed): Complexity classes and reductions, NP-completeness, nondeterminism, alternation, randomised computation, circuits, and subexponential-time complexity.
  • Game Theory – Ulle Endriss (period 5, completed): Noncooperative and cooperative game theory, mechanism design, and applications to strategic interaction and cooperation between rational agents.
  • Meaning, Reference and Modality – Tom Schoonen (period 7, in progress): Classical intensional and dynamic semantics from a philosophical and logical perspective, covering theories of meaning, reference, modality, and context through the work of Frege, Lewis, Stalnaker, and Kripke.