Search Talks

Bayesian causal structure learning aims to learn a posterior distribution over directed acyclic graphs (DAGs), and the mechanisms that define the relationship between parent and child variables. By taking a Bayesian approach, it is possible to reason about the uncertainty of the causal model. The notion of modelling the uncertainty over models is particularly crucial for causal structure learning since the model could be unidentifiable when given only a finite amount of observational data. In this paper, we introduce a novel method to jointly learn the structure and mechanisms of the causal model using Variational Bayes, which we call Variational Bayes-DAG-GFlowNet (VBG). We extend the method of Bayesian causal structure learning using GFlowNets to learn not only the posterior distribution over the structure, but also the parameters of a linear-Gaussian model. Our results on simulated data suggest that VBG is competitive against several baselines in modelling the posterior over DAGs and mechanisms, while offering several advantages over existing methods, including the guarantee to sample acyclic graphs, and the flexibility to generalize to non-linear causal mechanisms.

---

Zoom link 

I will argue that there are quantum states of the field theories of general relativity and electromagnetism that we typically ignore, but have interesting phenomenological effects. These states amount to relaxing the constraint equations known as the Hamiltonian and momentum constraints in GR and Gauss’ law in EM. Turning off the Hamiltonian constraint sources non-dynamical parts of the metric which mimic a pressure-less dust, and thus these effects may be the explanation as to why we have inferred the existence of dark matter, both locally and cosmologically. Turning off the momentum constraints add additional velocity-dependent source terms to this effective dust, but these effects are not conserved and redshift quickly outside the horizon. Turning off the Gauss’ law constraint mimics a charge density that does not respond to electric forces, but follows geodesics, thus adding a charged component to the dust. The effects in electromagnetism may have interesting impacts on BBN, the baryon-photon fluid during and after recombination, galactic dynamics, and cosmic rays. If this new structure in the gravitational and electric fields explain dark matter, it forbids an early period of inflation and therefore requires a different explanation for density perturbations.

We develop a Machine-Learning Renormalization Group (MLRG) algorithm to explore and analyze many-body lattice models in statistical physics. Using the representation learning capability of generative modeling, MLRG automatically learns the optimal renormalization group (RG) transformations from self-generated spin configurations and formulates RG equations without human supervision. The algorithm does not focus on simulating any particular lattice model but broadly explores all possible models compatible with the internal and lattice symmetries given the on-site symmetry representation. It can uncover the RG monotone that governs the RG flow, assuming a strong form of the $c$-theorem. This enables several downstream tasks, including unsupervised classification of phases, automatic location of phase transitions or critical points, controlled estimation of critical exponents, and operator scaling dimensions. We demonstrate the MLRG method in two-dimensional lattice models with Ising symmetry and show that the algorithm correctly identifies and characterizes the Ising criticality.

---

Zoom link