Four-Dimensional Universe Models Gain Stability with Three Potential Building Blocks

Scientists investigate the fundamental nature of quantum gravity using random tensor models as a tool to construct building blocks for spacetime. Alicia Castro, Astrid Eichhorn, and Razvan Gurau, all from the Institut f ür Theoretische Physik at the Universität Heidelberg, present a detailed analysis of order-4 random tensor models, seeking fixed points that describe the behaviour of four-dimensional spacetime at a fundamental level. Their research is significant because it challenges the established understanding of universality classes in quantum gravity, suggesting that simplified combinatorial models of spacetime and the well-known Reuter fixed point may belong to distinct categories. By optimising their calculations and identifying key parameter values, the authors reveal a phase portrait with only one stable fixed point possessing fewer relevant directions than expected, thereby offering new insights into the possible structures of quantum gravity.

Identifying potential fixed points in four-dimensional random tensor models with dynamical triangulations

Scientists are employing random tensor models as combinatorial tools to generate Euclidean dynamical triangulations, seeking a physical continuum limit analogous to the double-scaling limit observed in random matrices. This limit corresponds to a fixed point within a pregeometric Renormalization Group flow, where the tensor size functions as the Renormalization Group scale.
The current work focuses on identifying such fixed points within order-4 random tensor models associated with dynamical triangulations in four dimensions. Researchers investigated the resulting phase portrait as a function of regulator parameters, optimising results to pinpoint parameter values exhibiting minimal sensitivity to variations.

Analysis within a symmetric setting revealed three fixed-point candidates, though only one remains real across the entire parameter range and possesses only two relevant directions. This finding contrasts with the well-established Reuter fixed point in continuum quantum gravity, which is believed to be characterised by three relevant directions.
Consequently, the study suggests that simple combinatorial models of Euclidean triangulations and the Reuter fixed point likely belong to distinct universality classes. The research builds upon previous work introducing a Renormalization Group flow with respect to tensor size, proposing that critical regimes can be explored using functional RG techniques.

Specifically, the team investigated a random tensor model defined by a partition function involving a real tensor with four indices, transforming under the O(N)⊗4 group, and a polynomial invariant of degree eight in the tensor entries. Initial findings identified a fixed-point candidate with two relevant directions, but the possibility of a third could not be definitively excluded due to systematic uncertainties.

This work aims to address these uncertainties and rigorously determine the sign of the third critical exponent, crucial for establishing compatibility with the Reuter fixed point. To achieve this, researchers adapted a technique utilising a two-parameter family of regulators and identifying regions of minimal sensitivity where results remain stable despite parameter variations.

The functional RG flow was studied with respect to the tensor size N, reflecting the idea that RG flows should integrate out degrees of freedom, transitioning from microscopic to effective descriptions. The approach relies on examining the effective average action at scale N, which incorporates all quantum fluctuations at scales larger than N.

This effective action includes operators consistent with the theory’s symmetries and evolves across scales according to the Wetterich equation, a central equation in functional RG analysis. A diagonal regulator function was employed to suppress infrared modes, effectively providing a mass-like cutoff for tensor entries.

Renormalisation Group flow and fixed-point analysis of order-four tensor models

Random tensor models underpin this work as combinatorial tools for generating Euclidean dynamical triangulations. Researchers implemented a discrete, pregeometric variant of the Wetterich equation to model a Renormalization Group (RG) flow with respect to a scale counting tensor components rather than momentum modes.

This approach mirrors the RG flow in matrix size initially proposed by Brezin and Zinn-Justin, and builds upon prior studies of order-four, O(N)⊗4-symmetric tensor models. The study focused on identifying fixed points within these order-4 random tensor models associated with dynamical triangulations in four dimensions, seeking a physical continuum limit analogous to the double-scaling limit of random matrices.

To achieve this, the research team investigated the robustness of fixed-point candidates against changes in regulator parameters. They performed calculations within the largest truncation, identifying three fixed-point candidates, discarding two based on their Euclidean distance from a previously identified fixed point in a smaller truncation.

Detailed analysis revealed that the initially selected fixed point, designated A, was not real across the entire range of regulator parameters and exhibited only two relevant directions. Further investigation uncovered a “partner” fixed point, B, which collided with fixed point A and possessed three relevant directions when real.

A third fixed point, C, remained real across all parameter values and consistently displayed two critical exponents with positive real parts. Researchers examined both vanishing and cutoff-dominated regimes, confirming the robustness of fixed point C and identifying a point of minimal sensitivity to one regulator parameter. Through these meticulous calculations, the study aimed to determine whether these fixed points align with the Reuter universality class of continuum asymptotic safety, ultimately concluding that fixed point A is unlikely to correspond to this class.

Functional Renormalization Group analysis of order-4 tensor models reveals three fixed-point candidates

Random tensor models generate random geometries via Feynman diagrams dual to discretizations of four-dimensional spaces. Investigations into these models aim to identify a universal continuum limit potentially mirroring the Reuter fixed point in quantum gravity. The study focuses on order-4 random tensor models, specifically examining a partition function Z defined by a real tensor with four indices transforming under the O(N)⊗4 group.

Initial analyses identified a fixed-point candidate with two relevant directions, but the presence of a third could not be definitively excluded. This work builds upon previous research by employing functional Renormalization Group techniques to assess the robustness of these findings and determine the sign of the third critical exponent.

Researchers utilised a two-parameter family of regulators, searching for regions of minimal sensitivity where results remain consistent despite parameter variations. Three fixed-point candidates were discovered during the analysis of the phase portrait, however, only one remains real across the entire parameter range.

This real fixed point exhibits only two relevant directions, a key distinction from the Reuter fixed point which is believed to possess three. Detailed analysis revealed that the identified fixed-point candidate consistently displays two relevant directions, irrespective of the regulator parameters employed within the minimal sensitivity regions.

The asymptotic behaviour in relation to the parameter α further confirmed this observation. Consequently, the research concludes that simple combinatorial models of Euclidean triangulations, as explored in this study, and the Reuter fixed point likely belong to different universality classes. This suggests a divergence in their underlying critical behaviour and scaling properties.

Fixed point analysis of order-4 random tensor models and deviations from the Reuter fixed point

Researchers investigated random tensor models as a means of generating Euclidean dynamical triangulations and exploring potential connections to quantum gravity. These models utilise tensors to represent discrete geometries, with the aim of identifying a continuum limit analogous to those found in matrix models and potentially replicating the behaviour of the Reuter fixed point in four-dimensional quantum gravity.

The study employed functional renormalization group techniques to analyse the behaviour of order-4 random tensor models, focusing on identifying fixed points within a specific parameter space. Analysis revealed three fixed-point candidates, though only one remained real across the entire parameter range examined.

Importantly, this candidate exhibited only two relevant directions, differing significantly from the Reuter fixed point which is believed to possess three. This discrepancy suggests that the simple combinatorial models of Euclidean triangulations studied here likely belong to a different universality class than the Reuter fixed point.

The authors acknowledge limitations related to the specific model and parameter choices investigated, and propose further research focusing on exploring a wider range of tensor models and regulator parameters to potentially uncover fixed points with more relevant directions. Future work could also investigate the impact of different symmetry assumptions on the resulting phase portrait and fixed-point structure.