Physics · InformaticsTheory
Researcher in physics & quantum information
I’m fascinated by the interplay between physics and computation. My research brings together quantum computation, fundamental physics, and algorithmic information.
I develop the explanatory power of Heisenberg-picture descriptors for quantum protocols, Bell correlations, and the emergence of classicality.
In the Heisenberg picture of unitary quantum theory, Bell inequalities are violated with local elements of reality interacting locally. I analyze the comparison of measurement records as a quantum interaction, showing how the branching structure of the joint record produces Bell correlations.
I unify three fully local formalisms of quantum theory and quantify their descriptive cost. The dimension of a single qubit’s space of local descriptions grows exponentially with the total number of qubits—a cost I show is expected for a local and complete account.
Using the Heisenberg picture, Deutsch and Hayden traced the flow of the apparently teleported state parameters through the two classical bits. I further develop this account, showing how the information survives decoherence and propagates through the local chain reactions of what we call classical communication.
Contrary to conventional wisdom, Deutsch and Hayden proved that quantum systems admit complete local descriptions—even when entangled. I give a step-by-step guide to their descriptors, explaining how and why they work.
The Schrödinger and Heisenberg pictures make identical predictions, but are not equivalent descriptions of reality: many distinct sets of Heisenberg descriptors correspond to the same wave function. Their additional structure enables fully local explanations of superdense coding, teleportation, branching, and Bell correlations.
When does one quantum system influence another? For unitary dynamics, I show how the simple requirement of functional independence between Heisenberg descriptors recovers the more intricate no-influence criterion used in quantum causal models.
Rooted in computation rather than probability, algorithmic information theory quantifies the information intrinsic to an individual object. This objective approach to information offers a framework for studying thermodynamics and the nontrivial organization of complex physical systems.
With Sophie Berthelette, Xavier Coiteux-Roy, and Stefan Wolf
Statistical mechanics treats entropy as fundamental, yet its unit, joules per kelvin, depends on that of temperature—even though temperature is itself defined from entropy. We resolve this circularity by giving entropy its own unit—the bit—and recover Gibbs's duality between free energy and available information capacity.
With Geoffroy Bergeron
We propose a quantitative definition of emergence based on multiple drops in the Kolmogorov structure function of observational data. These drops mark sharp gains in compression as models become more complex. We apply the framework to dynamical systems and thermodynamics.
Sophistication and busy-beaver depth are essentially equivalent measures of nontrivial organization. But when the measures are relativized to auxiliary information, that information’s ability to solve the halting problem introduces a distortion between them.
Proceedings of the Royal Society A, 481, 20250553.
American Journal of Physics, 93(5), 390–395.
Physics, 7(2), 11.
IEEE International Conference on Quantum Computing and Engineering (QCE), 516–517.
Quantum Reports, 5(2), 510–525.
Entropy, 24(7), 985.
Quantum Reports, 3(2), 272–285.
Proceedings of the Royal Society A, 477(2246), 20200602.
arXiv, 2002.06709.
A presentation of a local account of Bell correlations.
Watch the talkA crash course on Deutsch–Hayden descriptors.
Watch the talkA talk accompanying Relativity of Depth and Sophistication.
Watch the talk
7 minA seven-minute sketch of causal reasoning with Heisenberg-picture descriptors.
Watch the talkI moderated this conversation between David Deutsch and Charles H. Bennett for my course on algorithmic information theory.
Watch the conversationA closer look at what happens in quantum teleportation.
Watch the talkThe CWI seminar presentation.
Watch the talk
A talk on local descriptions of quantum systems and the information they require.
Watch the talkMy contributions to two books edited by Alyssa Ney and Logan Chipkin.
Cover · Oxford University PressIn Local Quantum Mechanics: Everett, Many Worlds, and Reality
Edited by Alyssa Ney · Oxford University Press
The Schrödinger and Heisenberg pictures make identical predictions, but are not equivalent descriptions of reality: many distinct sets of Heisenberg descriptors correspond to the same wave function. Their additional structure enables fully local explanations of superdense coding, teleportation, branching, and Bell correlations.
Cover · Conjecture PressIn Bold Conjectures: Volume II · Essays Across Physics
Edited by Logan Chipkin · Conjecture Press
Podcasts
Two conversations will soon be available here.
02 / Teaching
Teaching is sharing the beauty of ideas and an obsession with problems—then discovering, in their solutions, even better ones.
“Without exaggeration, the best course I've ever taken. […] The interest you sparked in me for quantum computing convinced me to pursue graduate studies and explore the field further. Thank you so much, Charles.”
Algorithmic information theory measures the information in an individual object by the length of the shortest program that produces it. This self-contained graduate course begins with codes and Shannon information, then reviews computability theory: universal Turing machines, the halting problem, and Rice’s theorem.
From these foundations, we develop plain and prefix complexity, connect randomness with incompressibility, and explore Solomonoff’s induction. Chaitin’s halting probability Ω and incompleteness theorem reveal quantitative limits on what formal axiomatic systems can prove. We then study the coding theorem, which links complexity to universal probability, and use lower semicomputable semimeasures to establish the chain rule.
My exposition moves freely between computer science, mathematics, physics, and philosophy. Originally taught at Università della Svizzera italiana in spring 2023, the course is freely available as a lecture series accompanied by notes and problem sheets.
I moderated this conversation between David Deutsch and Charles H. Bennett for my course on algorithmic information theory.
Watch the conversation03 / Service
10–11 November 2026
Upcoming
Co-organizer, host and speaker
Université de Montréal · Campus MIL
2025–present
Founder and organizer
A continuing research seminar with local and international speakers.
11–15 August 2026
Co-organizer and public-day host
A five-day international workshop with FQxI.
2023
Co-organizer of the event and website
With Artur Ekert, Chiara Marletto and Samuel Kuypers.
2021
Co-organizer with Samuel Kuypers
An international conference on unitary quantum theory.
04 / How I approach science
Science is too important not to have fun doing it.
The meeting of ideas and the search for better explanations are what bring me back to work each day—if we can call this work.
For research conversations, collaborations and invitations.