Lídia Pacheco Cañamero B. del Rio: Catalogue data in Spring Semester 2022 
Name  Dr. Lídia Pacheco Cañamero B. del Rio 
Name variants  Lídia del Rio Lidia del Rio 
Department  Physics 
Relationship  Lecturer 
Number  Title  ECTS  Hours  Lecturers  

402046200L  Advanced Topics in Quantum Information Theory  8 credits  3V + 1U  L. Pacheco Cañamero B. del Rio, R. Silva  
Abstract  Solid introduction on advanced topics in quantum information theory, including: quantum thermodynamics, quantum clocks and control, measurement theory, quantum learning theory and quantum foundations. Prerequisites: Quantum Information Theory or equivalent courses.  
Learning objective  To prepare master students for a PhD or industry career by providing a selection of active research topics in quantum information theory and related areas.  
Content  1. Quantum thermodynamics a) Virtual qubits, virtual temperatures b) Qubit swaps c) Passivity and complete passivity d) Equilibration, Jaynes principle, thermal states and baths e) Resource theories: noisy operations, majorization and entropy f) Resource theories: thermal operations, thermal majorization and free energy g) Maxwell’s demon and Szilard’s engine, Landauer erasure h) Thermodynamics protocols for finitesize systems i) Autonomous thermal machines: master equation, continuous dynamics, steady states j) Autonomous thermal machines: types of engines, working regimes 2. Clocks and control a) Ideal quantum clocks b) Quasiideal clocks c) Informationtheoretical analysis 3. Puzzles and nogo theorems a) Hardy’s experiment (setup, simplified version, logical analysis) b) Quantum pigeonhole experiment (setup, simplified version, logical analysis) c) Physical implementation of measurements (von Neumann measurement scheme, strong and weak measurements, weak values) d) Replacing counterfactuals with weak measurements (Hardy and pigeonhole experiments) e) Replacing counterfactuals with measurements by different agents (FrauchigerRenner experiment) f) Pre and postselection paradoxes: definition and example g) Contextuality: operational definition and relation to paradoxes 4. Quantum learning theory (guest lecturer Marco Tomamichel) Quantum learning theory provides the theoretical foundations for machine learning involving quantum objects, where the quantum aspect can either come from the learner itself (e.g. quantum algorithms for machine learning) or the object to be studied (e.g. state tomography), or both. Quantum information theory tools can establish fundamental limits for such learning tasks. We will in particular explore applications of information theory to the following learning tasks: a) Sampleoptimal learning of quantum states b) Quantum PAC learning c) Multiarmed quantum bandits  
Lecture notes  Provided for the majority of contents; handwritten lecturer notes for the rest.  
Literature  Selected papers will be recommended to read throughout the semester. For example, for the quantum learning part: [1] Haah et al., Sampleoptimal tomography of quantum states, arXiv:1508.01797. [2] Arunachalam and de Wolf, Optimal Quantum Sample Complexity of Learning Algorithms, arXiv:1607.00932. [3] Lumbreras et al., Multiarmed quantum bandits: Exploration versus exploitation when learning properties of quantum states, arXiv:2108.13050.  
Prerequisites / Notice  Quantum Information Theory or equivalent course is necessary. Students should be familiar with density matrices, quantum channels (TPCPMs), Hamiltonian evolution and partial trace. Familiarity with quantum entropy measures helps but is not strictly necessary.  
Competencies 
