arXiv subjects

Antonio Rodríguez-Sánchez

Publications and source records attributed to Antonio Rodríguez-Sánchez.

At least 19 recordsLinked to original sources

arXiv2026

Probing baryon number with missing energy

Quark portal interactions $qqqN$ with a light singlet fermion $N$ make baryon number testable through missing transverse energy (MET). We find that present LHC data constrain scales up to 10 TeV (MET plus jet), 8 TeV (MET plus top) and 11 TeV (MET plus bjet). With increasing mass, or larger portal couplings, $N$ becomes less long-lived, and gives clean displaced vertex signatures, which encourage dedicated searches. We also narrow down viable mesogenesis models with color triplet scalars to a mass range $\sim y \,\cdot 3 \, \text{TeV}$ (with charm) and $\sim y \, \cdot 5 \, \text{TeV}$ (charmless) couplings $y$ to $b$ and lighter quarks, a window that can be scrutinized by HL-LHC. Interactions also induce rare decays of type baryon (meson) to meson (baryon) plus invisible, which complement high-$p_T$ searches and can prove baryon number violation. We explore charm decays $Λ_c \to (π,K) + \mathrm{invisible}$. Their branching ratios are subject to sizable hadronic uncertainties and require high luminosity flavor facilities such as a Tera-Z facility (FCC-ee, CEPC). Branching ratios of top quarks into one or two $b$-jets plus $N$ can reach few$\times 10^{-6}$.

hep-ph↗

arXiv2026

Scalable and Efficient Continual Learning from Demonstration via a Hypernetwork-generated Stable Dynamics Model

Robots capable of learning from demonstration (LfD) must exhibit stability while executing learned motion skills. To be effective in the real world, they should also remember multiple skills over time -- a capability lacking in current stable-LfD methods. We propose an approach to stable, continual LfD, and highlight the role of stability in improving continual learning. Our proposed hypernetwork generates the parameters of two neural networks: a trajectory learning dynamics model, and a trajectory-stabilizing Lyapunov function. These generated networks form a clock-augmented stable neural ODE solver (sNODE), a stable dynamics model that offers a superior stability-accuracy trade-off compared to the state-of-the-art. We further propose stochastic hypernetwork regularization with a single, uniformly-sampled task embedding, reducing the cumulative training time for $N$ tasks from O($N^2$) to O($N$) without degrading performance on real-world tasks. We introduce high-dimensional variants of the popular LASA dataset to assess scalability and extend a dataset of robotic LfD tasks to assess real-world performance. We empirically evaluate our approach on multiple LfD datasets of varying complexity, including sequences of 7--26 tasks, trajectories of 2--32 dimensions, and real-world tasks involving position and orientation. Our thorough evaluation on multiple LfD datasets demonstrates that our approach sequentially learns and retains multiple motion skills without retraining on past demonstrations, and outperforms other relevant baselines in terms of trajectory errors, continual learning scores, and stability metrics. Notably, we show that stability greatly enhances continual learning performance, particularly in size-efficient chunked hypernetworks. Our code is available at https://github.com/sayantanauddy/clfd-snode.

cs.RO↗

arXiv2026

On the Charm Contribution to the Muon $g-2$ Light-by-Light

We combine existing perturbative results to show that a precise analytic determination of the charm-quark contribution to the hadronic light-by-light (HLbL) part of the muon anomalous magnetic moment is possible. Working in the $\overline{\mathrm{MS}}$ scheme, we include the NLO $\mathcal{O}(α_s)$ correction, which significantly reduces the residual renormalization-scale dependence and the perturbative uncertainty. Our final result is $a_μ^{\mathrm{HLbLc}}=3.65(25)\times 10^{-11}$, in good agreement with recent lattice determinations.

hep-ph↗

arXiv2026

The Pion Sum Rule beyond the Chiral Limit

The difference between the charged and neutral pion masses can be predicted from a well-known dispersion relation involving an infinite-energy integral over experimental data, the pion sum rule. This relation, however, holds only in the chiral limit. Here we combine several nonperturbative techniques to determine the form of the physical finite-energy integral, thereby generalizing this sum rule to include effects linear in the light-quark masses. We test the dispersion relation using hadronic tau-decay data.

hep-ph↗

arXiv2025

Dynamic Sparsity: Challenging Common Sparsity Assumptions for Learning World Models in Robotic Reinforcement Learning Benchmarks

The use of learned dynamics models, also known as world models, can improve the sample efficiency of reinforcement learning. Recent work suggests that the underlying causal graphs of such dynamics models are sparsely connected, with each of the future state variables depending only on a small subset of the current state variables, and that learning may therefore benefit from sparsity priors. Similarly, temporal sparsity, i.e. sparsely and abruptly changing local dynamics, has also been proposed as a useful inductive bias. In this work, we critically examine these assumptions by analyzing ground-truth dynamics from a set of robotic reinforcement learning environments in the MuJoCo Playground benchmark suite, aiming to determine whether the proposed notions of state and temporal sparsity actually tend to hold in typical reinforcement learning tasks. We study (i) whether the causal graphs of environment dynamics are sparse, (ii) whether such sparsity is state-dependent, and (iii) whether local system dynamics change sparsely. Our results indicate that global sparsity is rare, but instead the tasks show local, state-dependent sparsity in their dynamics and this sparsity exhibits distinct structures, appearing in temporally localized clusters (e.g., during contact events) and affecting specific subsets of state dimensions. These findings challenge common sparsity prior assumptions in dynamics learning, emphasizing the need for grounded inductive biases that reflect the state-dependent sparsity structure of real-world dynamics.

cs.LG↗

arXiv2025

Constraints on baryon-number-violating top-quark operators in standard model effective field theory

Within the Standard Model Effective Field Theory framework, we set indirect constraints on top quark operators that violate baryon number by one unit above the TeV scale. We find that these constraints are typically many orders of magnitude more stringent than the recently derived direct bounds from collider experiments. Therefore, direct observation of baryon number violation in these top-quark observables at the TeV scale would imply a large fine-tuning among operators across different energy scales. This possibility is not protected under universal radiative corrections or any known symmetry principles.

hep-ph↗

arXiv2025

Hadronic tau decays

We give a pedagogical introduction to the rich phenomenology of hadronic tau decays. These decays provide a unique window into the interplay of electroweak and strong interactions at low energies, as they occur primarily via $W$ exchange after the electroweak quark current hadronizes. In this manuscript, we summarize the basic ingredients required to perform precision physics studies in this sector. We detail the derivation of the different distributions within the Standard Model, discuss how to parametrize the non-perturbative QCD dynamics, and present methods commonly used to achieve clean theoretical predictions. Additionally, we briefly review how these distributions generalize in the presence of relatively heavy particles from beyond the Standard Model. This overview thus aims to serve as a useful starting point for readers interested in understanding how the only lepton capable of decaying into hadrons does so.

hep-ph↗

arXiv2025

Short-distance contributions to Hadronic-light-by-light for the muon $g-2$

This talk discusses short-distance contributions to the hadronic light-by-light part of the muon g-2 as in the Standard Model. A short discussion about the theory prediction is followed by the status of our work. The main new results since the previous chiral dynamics workshop is how to calculate the higher order corrections to the Melnikov-Vainshtein region using the operator product expansion. We work out fully the next order in the OPE, $D=4$, including photon and gluon only operators. In addition we briefly discuss the ongoing work on gluonic corrections to the next-order and nonperturbative estimates.

hep-ph↗

arXiv2025

Constraints on the hadronic light-by-light tensor in corner kinematics for the muon $g-2$

The dispersive approach to the hadronic light-by-light contribution to the muon $g-2$ involves an integral over three virtual photon momenta appearing in the light-by-light tensor. Building upon previous works, we systematically derive short-distance constraints in the region where two momenta are large compared to the third, the so-called Melnikov-Vainshtein or corner region. We include gluonic corrections for the different scalar functions appearing in the Lorentz decomposition of the underlying tensor, and explicitly check analytic agreement with alternative operator product expansions in overlapping regimes of validity. A very strong pattern of cancellations is observed for the final $g-2$ integrand. The last observation suggests that a very compact expression only containing the axial current form factors can provide a good approximation of the corner region of the hadronic light-by-light tensor.

hep-ph↗

arXiv2024

Vision Transformers for Weakly-Supervised Microorganism Enumeration

Microorganism enumeration is an essential task in many applications, such as assessing contamination levels or ensuring health standards when evaluating surface cleanliness. However, it's traditionally performed by human-supervised methods that often require manual counting, making it tedious and time-consuming. Previous research suggests automating this task using computer vision and machine learning methods, primarily through instance segmentation or density estimation techniques. This study conducts a comparative analysis of vision transformers (ViTs) for weakly-supervised counting in microorganism enumeration, contrasting them with traditional architectures such as ResNet and investigating ViT-based models such as TransCrowd. We trained different versions of ViTs as the architectural backbone for feature extraction using four microbiology datasets to determine potential new approaches for total microorganism enumeration in images. Results indicate that while ResNets perform better overall, ViTs performance demonstrates competent results across all datasets, opening up promising lines of research in microorganism enumeration. This comparative study contributes to the field of microbial image analysis by presenting innovative approaches to the recurring challenge of microorganism enumeration and by highlighting the capabilities of ViTs in the task of regression counting.

cs.CV↗

arXiv2024

Constraints on subleading interactions in beta decay Lagrangian

We discuss the effective field theory (EFT) for nuclear beta decay. The general quark-level EFT describing charged-current interactions between quarks and leptons is matched to the nucleon-level non-relativistic EFT at the O(MeV) momentum scale characteristic for beta transitions. The matching takes into account, for the first time, the effect of all possible beyond-the-Standard-Model interactions at the subleading order in the recoil momentum. We calculate the impact of all the Wilson coefficients of the leading and subleading EFT Lagrangian on the differential decay width in allowed beta transitions. As an example application, we show how the existing experimental data constrain the subleading Wilson coefficients corresponding to pseudoscalar, weak magnetism, and induced tensor interactions. The data display a 3.5 sigma evidence for nucleon weak magnetism, in agreement with the theory prediction based on isospin symmetry.

hep-ph↗

arXiv2023

Continual Learning from Demonstration of Robotics Skills

Methods for teaching motion skills to robots focus on training for a single skill at a time. Robots capable of learning from demonstration can considerably benefit from the added ability to learn new movement skills without forgetting what was learned in the past. To this end, we propose an approach for continual learning from demonstration using hypernetworks and neural ordinary differential equation solvers. We empirically demonstrate the effectiveness of this approach in remembering long sequences of trajectory learning tasks without the need to store any data from past demonstrations. Our results show that hypernetworks outperform other state-of-the-art continual learning approaches for learning from demonstration. In our experiments, we use the popular LASA benchmark, and two new datasets of kinesthetic demonstrations collected with a real robot that we introduce in this paper called the HelloWorld and RoboTasks datasets. We evaluate our approach on a physical robot and demonstrate its effectiveness in learning real-world robotic tasks involving changing positions as well as orientations. We report both trajectory error metrics and continual learning metrics, and we propose two new continual learning metrics. Our code, along with the newly collected datasets, is available at https://github.com/sayantanauddy/clfd.

cs.RO↗

arXiv2023

Constraints on the hadronic light-by-light in the Melnikov-Vainshtein regime

The muon anomalous magnetic moment continues to attract attention due to the possible tension between the experimentally measured value and the theoretical Standard Model prediction. With the aim to reduce the uncertainty on the hadronic light-by-light contribution to the magnetic moment, we derive short-distance constraints in the Melnikov-Vainshtein regime which are useful for data-driven determinations. In this kinematical region, two of the four electromagnetic currents are close in the four-point function defining the hadronic light-by-light tensor. To obtain the constraints, we develop a systematic operator product expansion of the tensor in question to next-to-leading order in the expansion in operators. We evaluate the leading in $α_s$ contributions and derive constraints for the next-to-leading operators that are also valid nonperturbatively.

hep-ph↗

arXiv2022

Affordance detection with Dynamic-Tree Capsule Networks

Affordance detection from visual input is a fundamental step in autonomous robotic manipulation. Existing solutions to the problem of affordance detection rely on convolutional neural networks. However, these networks do not consider the spatial arrangement of the input data and miss parts-to-whole relationships. Therefore, they fall short when confronted with novel, previously unseen object instances or new viewpoints. One solution to overcome such limitations can be to resort to capsule networks. In this paper, we introduce the first affordance detection network based on dynamic tree-structured capsules for sparse 3D point clouds. We show that our capsule-based network outperforms current state-of-the-art models on viewpoint invariance and parts-segmentation of new object instances through a novel dataset we only used for evaluation and it is publicly available from github.com/gipfelen/DTCG-Net. In the experimental evaluation we will show that our algorithm is superior to current affordance detection methods when faced with grasping previously unseen objects thanks to our Capsule Network enforcing a parts-to-whole representation.

cs.CV↗

arXiv2022

Short-distance constraints on the hadronic light-by-light

The muon anomalous magnetic moment continues to attract interest due to the potential tension between experimental measurement [1,2] and the Standard Model prediction [3]. The hadronic light-by-light contribution to the magnetic moment is one of the two diagrammatic topologies currently saturating the theoretical uncertainty. With the aim of improving precision on the hadronic light-by-light in a data-driven approach founded on dispersion theory [4,5], we derive various short-distance constraints of the underlying correlation function of four electromagnetic currents. Here, we present our previous progress in the purely short-distance regime and current efforts in the so-called Melnikov-Vainshtein limit.

hep-ph↗

arXiv2022

Momentum Capsule Networks

Capsule networks are a class of neural networks that achieved promising results on many computer vision tasks. However, baseline capsule networks have failed to reach state-of-the-art results on more complex datasets due to the high computation and memory requirements. We tackle this problem by proposing a new network architecture, called Momentum Capsule Network (MoCapsNet). MoCapsNets are inspired by Momentum ResNets, a type of network that applies reversible residual building blocks. Reversible networks allow for recalculating activations of the forward pass in the backpropagation algorithm, so those memory requirements can be drastically reduced. In this paper, we provide a framework on how invertible residual building blocks can be applied to capsule networks. We will show that MoCapsNet beats the accuracy of baseline capsule networks on MNIST, SVHN, CIFAR-10 and CIFAR-100 while using considerably less memory. The source code is available on https://github.com/moejoe95/MoCapsNet.

cs.CV↗

arXiv2022

On the sensitivity of the D parameter to new physics

Measurements of angular correlations in nuclear beta decay are important tests of the Standard Model (SM). Among those, the so-called D correlation parameter occupies a particular place because it is odd under time reversal, and because the experimental sensitivity is at the $10^{-4}$ level, with plans of further improvement in the near future. Using effective field theory~(EFT) techniques, we reassess its potential to discover or constrain new physics beyond the SM. We provide a comprehensive classification of CP-violating EFT scenarios which generate a shift of the D parameter away from the SM prediction. We show that, in each scenario, a shift larger than $10^{-5}$ is in serious tension with the existing experimental data, where bounds coming from electric dipole moments and LHC observables play a decisive role. The tension can only be avoided by fine tuning of the parameters in the UV completion of the EFT. We illustrate this using examples of leptoquark UV completions. Finally, we comment on the possibility to probe CP-conserving new physics via the D parameter.

hep-ph↗

arXiv2022

Improving the Trainability of Deep Neural Networks through Layerwise Batch-Entropy Regularization

Training deep neural networks is a very demanding task, especially challenging is how to adapt architectures to improve the performance of trained models. We can find that sometimes, shallow networks generalize better than deep networks, and the addition of more layers results in higher training and test errors. The deep residual learning framework addresses this degradation problem by adding skip connections to several neural network layers. It would at first seem counter-intuitive that such skip connections are needed to train deep networks successfully as the expressivity of a network would grow exponentially with depth. In this paper, we first analyze the flow of information through neural networks. We introduce and evaluate the batch-entropy which quantifies the flow of information through each layer of a neural network. We prove empirically and theoretically that a positive batch-entropy is required for gradient descent-based training approaches to optimize a given loss function successfully. Based on those insights, we introduce batch-entropy regularization to enable gradient descent-based training algorithms to optimize the flow of information through each hidden layer individually. With batch-entropy regularization, gradient descent optimizers can transform untrainable networks into trainable networks. We show empirically that we can therefore train a "vanilla" fully connected network and convolutional neural network -- no skip connections, batch normalization, dropout, or any other architectural tweak -- with 500 layers by simply adding the batch-entropy regularization term to the loss function. The effect of batch-entropy regularization is not only evaluated on vanilla neural networks, but also on residual networks, autoencoders, and also transformer models over a wide range of computer vision as well as natural language processing tasks.

cs.LG↗