arXiv subjects

Anthony Francis

Publications and source records attributed to Anthony Francis.

At least 55 records · Page 3Linked to original sources

arXiv2021

Approaching the master-field: Hadronic observables in large volumes

The master-field approach to lattice QCD envisions performing calculations on a small number of large-volume gauge-field configurations. Substantial progress has been made recently in the generation of such fields, and this must be joined with measurement strategies that take advantage of the large volume. In these proceedings, we describe how to compute simple hadronic quantities efficiently and estimate their errors in the master-field approach, i.e. by studying cross-correlations of observables on a single configuration. We discuss the scaling of the uncertainty with the volume and compare extractions based on momentum-projected and position-space two-point functions. The latter show promising results, already at intermediate volumes, but come with additional technical complexities such as a more complicated manifestation of boundary effects, which we also address.

hep-lat↗

arXiv2020

Master-field simulations of O(a)-improved lattice QCD: Algorithms, stability and exactness

In master-field simulations of lattice QCD, the expectation values of interest are obtained from a single or at most a few representative gauge-field configurations on very large lattices. If the light quarks are included, the generation of these fields using standard techniques is however challenging in view of various algorithmic instabilities and precision issues. Ways to overcome these problems are described here for the case of the O(a)-improved Wilson formulation of lattice QCD and the viability of the proposed measures is then checked in extensive simulations of the theory with 2+1 flavours of quarks.

hep-lat↗

arXiv2020

Long-Range Indoor Navigation with PRM-RL

Long-range indoor navigation requires guiding robots with noisy sensors and controls through cluttered environments along paths that span a variety of buildings. We achieve this with PRM-RL, a hierarchical robot navigation method in which reinforcement learning agents that map noisy sensors to robot controls learn to solve short-range obstacle avoidance tasks, and then sampling-based planners map where these agents can reliably navigate in simulation; these roadmaps and agents are then deployed on robots, guiding them along the shortest path where the agents are likely to succeed. Here we use Probabilistic Roadmaps (PRMs) as the sampling-based planner, and AutoRL as the reinforcement learning method in the indoor navigation context. We evaluate the method in simulation for kinematic differential drive and kinodynamic car-like robots in several environments, and on differential-drive robots at three physical sites. Our results show PRM-RL with AutoRL is more successful than several baselines, is robust to noise, and can guide robots over hundreds of meters in the face of noise and obstacles in both simulation and on robots, including over 5.8 kilometers of physical robot navigation. Video: https://youtu.be/xN-OWX5gKvQ

cs.RO↗

arXiv2019

Lattice QCD estimate of the quark-gluon plasma photon emission rate

We present a computation of the photon emission rate of the quark-gluon plasma from two-flavor lattice QCD at a temperature of 254 MeV, which follows up on the work presented in [1]. We perform a continuum extrapolation of the vector-current correlator, and consider a linear combination of the Lorentz indices corresponding to a UV-finite spectral function. To extract the spectral function from the lattice correlators, an ill-posed inverse problem, we model the spectral function with a Padé ansatz. We further constrain our analysis by simultaneously fitting data with different momenta. We present results for a multi-momentum fit including the three smallest momenta available from our lattice analysis.

hep-lat↗

arXiv2019

Evolving Rewards to Automate Reinforcement Learning

Many continuous control tasks have easily formulated objectives, yet using them directly as a reward in reinforcement learning (RL) leads to suboptimal policies. Therefore, many classical control tasks guide RL training using complex rewards, which require tedious hand-tuning. We automate the reward search with AutoRL, an evolutionary layer over standard RL that treats reward tuning as hyperparameter optimization and trains a population of RL agents to find a reward that maximizes the task objective. AutoRL, evaluated on four Mujoco continuous control tasks over two RL algorithms, shows improvements over baselines, with the the biggest uplift for more complex tasks. The video can be found at: \url{https://youtu.be/svdaOFfQyC8}.

cs.LG↗

arXiv2019

Testing the strength of the $\text{U}_A(1)$ anomaly at the chiral phase transition in two-flavour QCD

We study the thermal transition of QCD with two degenerate light flavours by lattice simulations using $\mathcal{O}(a)$-improved Wilson quarks. Particular emphasis lies on the pattern of chiral symmetry restoration, which we probe via the static screening correlators. On $32^3$ volumes we observe that the screening masses in transverse iso-vector vector and axial-vector channels become degenerate at the transition temperature. The splitting between the screening masses in iso-vector scalar and pseudoscalar channels is strongly reduced compared to the splitting at zero temperature and is actually consistent with zero within uncertainties. In this proceedings article we extend our studies to matrix elements and iso-singlet correlation functions. Furthermore, we present results on larger volumes, including first results at the physical pion mass.

hep-lat↗

arXiv2019

Learning Navigation Behaviors End-to-End with AutoRL

We learn end-to-end point-to-point and path-following navigation behaviors that avoid moving obstacles. These policies receive noisy lidar observations and output robot linear and angular velocities. The policies are trained in small, static environments with AutoRL, an evolutionary automation layer around Reinforcement Learning (RL) that searches for a deep RL reward and neural network architecture with large-scale hyper-parameter optimization. AutoRL first finds a reward that maximizes task completion, and then finds a neural network architecture that maximizes the cumulative of the found reward. Empirical evaluations, both in simulation and on-robot, show that AutoRL policies do not suffer from the catastrophic forgetfulness that plagues many other deep reinforcement learning algorithms, generalize to new environments and moving obstacles, are robust to sensor, actuator, and localization noise, and can serve as robust building blocks for larger navigation tasks. Our path-following and point-to-point policies are respectively 23% and 26% more successful than comparison methods across new environments. Video at: https://youtu.be/0UwkjpUEcbI

cs.RO↗

arXiv2018

The $H$ dibaryon from lattice QCD with SU(3) flavor symmetry

We show recent results from the Mainz group using $N_f = 2 + 1$ CLS ensembles generated at the $SU(3)$ symmetric point. Temporal correlation functions using two-baryon interpolating operators are calculated with the distillation method. In addition to the spin-0 operators relevant for studying the $H$ dibaryon, we added spin-1 operators to our basis, thereby extending our results to other flavor sectors. These preliminary results show a finite-volume energy below the $ΛΛ$ threshold. Further calculations are necessary to establish whether the $H$ dibaryon is bound at the physical point.

hep-lat↗

arXiv2018

Evidence for charm-bottom tetraquarks and the mass dependence of heavy-light tetraquark states from lattice QCD

We continue our study of heavy-light four-quark states and find evidence from lattice QCD for the existence of a strong-interaction-stable $I(J^P)=0(1^+)$ $ud\bar{c}\bar{b}$ tetraquark with mass in the range of 15 to 61 MeV below $\bar{D}B^*$ threshold. Since this range includes the electromagnetic $\bar{D}Bγ$ decay threshold, current uncertainties do not allow us to determine whether such a state would decay electromagnetically, or only weakly. We also perform a study at fixed pion mass, with NRQCD for the heavy quarks, simulating $qq^\prime \bar{b}^\prime \bar{b}$ and $q q^\prime \bar{b}^\prime\bar{b}^\prime$ tetraquarks with $q,\, q^\prime =ud$ or $\ell s$ and variable, unphysical $m_{b^\prime}$ in order to investigate the heavy mass-dependence of such tetraquark states. We find that the dependence of the binding energy follows a phenomenologically-expected form and that, though NRQCD breaks down before $m_{b^\prime}=m_c$ is reached, the results at higher $m_{b^\prime}$ clearly identify the $ud\bar{b}^\prime \bar{b}$ channel as the most likely to support a strong-interaction-stable tetraquark state at $m_{b^\prime}=m_c$. This observation serves to motivate the direct $ud\bar{c}\bar{b}$ simulation. Throughout we use dynamical $n_f=2+1$ ensembles with pion masses $m_π=$415, 299, and 164 MeV reaching down almost to the physical point, a relativistic heavy quark prescription for the charm quark, and NRQCD for the bottom quark(s).

hep-lat↗

arXiv2018

Dark Matter from Strong Dynamics: The Minimal Theory of Dark Baryons

As a simple model for dark matter, we propose a QCD-like theory based on $\rm{SU}(2)$ gauge theory with one flavor of dark quark. The model is confining at low energy and we use lattice simulations to investigate the properties of the lowest-lying hadrons. Compared to QCD, the theory has several peculiar differences: there are no Goldstone bosons or chiral symmetry restoration when the dark quark becomes massless; the usual global baryon number symmetry is enlarged to $\rm{SU}(2)_B$, resembling isospin; and baryons and mesons are unified together in $\rm{SU}(2)_B$ iso-multiplets. We argue that the lightest baryon, a vector boson, is a stable dark matter candidate and is a composite realization of the hidden vector dark matter scenario. The model naturally includes a lighter state, the analog of the $η^\prime$ in QCD, for dark matter to annihilate into to set the relic density via thermal freeze-out. Dark matter baryons may also be asymmetric, strongly self-interacting, or have their relic density set via $3 \to 2$ cannibalizing transitions. We discuss some experimental implications of coupling dark baryons to the Higgs portal.

hep-ph↗

arXiv2018

PRM-RL: Long-range Robotic Navigation Tasks by Combining Reinforcement Learning and Sampling-based Planning

We present PRM-RL, a hierarchical method for long-range navigation task completion that combines sampling based path planning with reinforcement learning (RL). The RL agents learn short-range, point-to-point navigation policies that capture robot dynamics and task constraints without knowledge of the large-scale topology. Next, the sampling-based planners provide roadmaps which connect robot configurations that can be successfully navigated by the RL agent. The same RL agents are used to control the robot under the direction of the planning, enabling long-range navigation. We use the Probabilistic Roadmaps (PRMs) for the sampling-based planner. The RL agents are constructed using feature-based and deep neural net policies in continuous state and action spaces. We evaluate PRM-RL, both in simulation and on-robot, on two navigation tasks with non-trivial robot dynamics: end-to-end differential drive indoor navigation in office environments, and aerial cargo delivery in urban environments with load displacement constraints. Our results show improvement in task completion over both RL agents on their own and traditional sampling-based planners. In the indoor navigation task, PRM-RL successfully completes up to 215 m long trajectories under noisy sensor conditions, and the aerial cargo delivery completes flights over 1000 m without violating the task constraints in an environment 63 million times larger than used in training.

cs.AI↗

arXiv2017

The pion quasiparticle in the low-temperature phase of QCD

We extend our previous studies [PhysRevD.90.054509, PhysRevD.92.094510] of the pion quasiparticle in the low-temperature phase of two-flavor QCD with support from chiral effective theory. This includes the analysis performed on a finite temperature ensemble of size $20\times 64^3$ at $T\approx 151$MeV and a lighter zero-temperature pion mass $m_π \approx 185$ MeV. Furthermore, we investigate the Gell-Mann--Oakes-Renner relation at finite temperature and the Dey-Eletsky-Ioffe mixing theorem at finite quark mass.

hep-lat↗

arXiv2017

More on heavy tetraquarks in lattice QCD at almost physical pion mass

We report on our progress in studying exotic, heavy tetraquark states, $qq\prime \bar Q\bar Q\prime$. Using publicly available dynamical $n_f =2+1$ Wilson-Clover gauge configurations, generated by the PACS-CS collaboration, with pion masses $\simeq$164, 299 and 415 MeV, we extend our previous analysis to heavy quark components containing heavier than physical bottom quarks $\bar Q\bar Q\prime=\bar b\prime\bar b\prime$ or $\bar Q\bar Q\prime=\bar b\bar b\prime$, charm and bottom quarks $\bar c\bar b$ and also only charm quarks $\bar c\bar c$. Throughout we employ NRQCD and relativistic heavy quarks for the heavier than bottom, bottom and charm quarks. Using our previously established diquark-antidiquark and meson-meson operator basis we comment in particular on the dependence of the binding energy on the mass of the heavy quark component $\bar Q\bar Q$, with heavy quarks ranging from $m_Q=0.85\ldots 6.3\cdot m_b$. In the heavy flavor non-degenerate case, $\bar Q\bar Q\prime$, and especially for the tetraquark channel $ud\bar c\bar b$, we extend our work to utilize a $3\times 3$ GEVP to study the ground and threshold states thereby enabling a clear identification of possible binding. Finally, we present initial work on the $\bar Q\bar Q\prime=\bar c\bar c$ system where a much larger operator basis is available in comparison to flavor combinations with NRQCD quarks.

hep-lat↗

arXiv2017

An estimate for the thermal photon rate from lattice QCD

We estimate the production rate of photons by the quark-gluon plasma in lattice QCD. We propose a new correlation function which provides better control over the systematic uncertainty in estimating the photon production rate at photon momenta in the range πT/2 to 2πT. The relevant Euclidean vector current correlation functions are computed with $N_{\mathrm f}$ = 2 Wilson clover fermions in the chirally-symmetric phase. In order to estimate the photon rate, an ill-posed problem for the vector-channel spectral function must be regularized. We use both a direct model for the spectral function and a model-independent estimate from the Backus-Gilbert method to give an estimate for the photon rate.

hep-lat↗

arXiv2017

Lattice Prediction for Deeply Bound Doubly Heavy Tetraquarks

We investigate the possibility of $qq^\prime \bar b \bar b$ tetraquark bound states using $n_f=2+1$ lattice QCD ensembles with pion masses $\simeq 164$, $299$, and $415$ MeV. Motivated by observations from heavy baryon phenomenology, we consider two lattice interpolating operators both of which are expected to couple efficiently to tetraquark states: one with diquark-antidiquark and one with a meson-meson structure. Using nonrelativistic QCD to simulate the bottom quarks, we study the $ud\bar b \bar b$, $\ell s\bar b \bar b$ channels with $\ell=u,d$, and find unambiguous signals for strong-interaction-stable $J^P=1^+$ tetraquarks. These states are found to lie $189(10)$ and $98(7)$ MeV below the corresponding free two-meson thresholds.

hep-lat↗

arXiv2017

On the strength of the $U_A(1)$ anomaly at the chiral phase transition in $N_f=2$ QCD

We study the thermal transition of QCD with two degenerate light flavours by lattice simulations using $O(a)$-improved Wilson quarks. Temperature scans are performed at a fixed value of $N_t = (aT)^{-1}=16$, where $a$ is the lattice spacing and $T$ the temperature, at three fixed zero-temperature pion masses between 200 MeV and 540 MeV. In this range we find that the transition is consistent with a broad crossover. As a probe of the restoration of chiral symmetry, we study the static screening spectrum. We observe a degeneracy between the transverse isovector vector and axial-vector channels starting from the transition temperature. Particularly striking is the strong reduction of the splitting between isovector scalar and pseudoscalar screening masses around the chiral phase transition by at least a factor of three compared to its value at zero temperature. In fact, the splitting is consistent with zero within our uncertainties. This disfavours a chiral phase transition in the $O(4)$ universality class.

hep-lat↗

arXiv2016

Static and non-static vector screening masses

Thermal screening masses of the conserved vector current are calculated both in a weak-coupling approach and in lattice QCD. The inverse of a screening mass can be understood as the length scale over which an external electric field is screened in a QCD medium. The comparison of screening masses both in the zero and non-zero Matsubara frequency sectors shows good agreement of the perturbative and the lattice results. Moreover, at $T\approx 508\mathrm{MeV}$ the lightest screening mass lies above the free result ($2πT$), in agreement with the $\mathcal{O}(g^2)$ weak-coupling prediction.

hep-lat↗

arXiv2016

Dark matter from one-flavor SU(2) gauge theory

SU(2) gauge theory with a single fermion in the fundamental representation is a minimal non-Abelian candidate for the dark matter sector, which is presently missing from the standard model. Having only a single flavor provides a natural mechanism for stabilizing dark matter on cosmological timescales. Preliminary lattice results are presented and discussed in the context of dark matter phenomenology.

hep-lat↗