International Centre for Theoretical Physics

The conformal bootstrap studies strongly coupled quantum field theories through symmetry, unitarity and crossing symmetry without relying on perturbation theory. This talk reviews recent analytic functional approaches to the numerical bootstrap and introduces new results. It then considers how machine learning could navigate consistent conformal field theories. Neural networks can construct crossing-symmetric correlators from limited physical input; their tendency toward simple, smooth solutions may reveal additional organizing principles behind bootstrap constraints.

Amino acids have long been used in biological and pharmaceutical formulations, but their stabilizing effects are often described only for particular proteins. This talk examines a broader mechanism: small molecules modify colloidal interactions in proteins, DNA and synthetic nanoparticles. Measurements of osmotic virial coefficients and potentials of mean force reveal consistent effects across these systems. A model of weakly interacting amino acids binding to patchy colloidal surfaces predicts experimentally measured dissociation constants; proline–lysozyme predictions and measurements agree within a factor of two. Experiments also test predictions that short peptides outperform the same number of free amino acids and that charged amino acids stabilize oppositely charged proteins. The consequences extend to suppression of liquid–liquid phase separation and cellular stress granules, and doubled insulin bioavailability in mice when co-formulated with proline. The results motivate treating amino acids as active regulators of colloidal behavior alongside familiar influences such as salt.

Debashree Ghosh discusses quantum-chemistry approaches for strongly correlated polyaromatic hydrocarbons, including acenes and carotenoids. The talk examines why large valence active spaces are needed to describe optical properties and singlet–triplet gaps, and how density-matrix renormalization-group methods address these systems. Changes in molecular topology and odd-membered rings are connected to spin frustration and optical behaviour. The seminar also covers neural-network configuration interaction, connections between artificial neural networks and matrix-product-state wavefunctions, and restricted-Boltzmann-machine approaches to quantum wavefunctions. Applications include singlet fission and multiexcitonic phenomena. The hybrid event takes place in ICTP’s Euler Lecture Hall and online; the official event page provides the registration link.