arXiv subjects

Lorenzo Mauro

Publications and source records attributed to Lorenzo Mauro.

7 recordsLinked to original sources

arXiv2026

Thermal effective action for the $O(N)$ vector model

We compute the leading coefficients of the thermal effective action for the critical O(N) vector model in three dimensions, in the large-N limit in presence of non vanishing angular twist. At high temperature, the partition function on a product of a two-dimensional spatial manifold and a thermal circle admits a Kaluza-Klein reduction to a local effective action on the spatial slice, whose coefficients encode universal CFT data such as the Casimir energy and the response to a Kaluza-Klein gauge field. We determine these coefficients through two independent computations: an evaluation of the twisted partition function on the two-sphere in the high-temperature limit, and a direct path-integral computation on a generic weakly curved background. The two methods yield consistent results, providing a non-trivial check of the thermal effective action framework.

hep-th↗

arXiv2025

Hole spin qubits in unstrained Germanium layers

Strained germanium heterostructures are one of the most promising material for hole spin qubits but suffer from the strong anisotropy of the gyromagnetic factors that hinders the optimization of the magnetic field orientation. The figures of merit (Rabi frequencies, lifetimes...) can indeed vary by an order of magnitude within a few degrees around the heterostructure plane. We propose to address this issue by confining the holes at the interface of an unstrained, bulk Ge substrate or thick buffer. We model such structures and show that the gyromagnetic anisotropy is indeed considerably reduced. In addition, the Rabi frequencies and quality factors can be significantly improved with respect to strained heterostructures. This extends the operational range of the qubits and shall ease the scale-up to many-qubit systems.

cond-mat.mes-hall↗

arXiv2025

Strain engineering in Ge/GeSi spin qubits heterostructures

The heavy-holes in Ge/GeSi heterostructures show highly anisotropic gyromagnetic response with in-plane $g$-factors $g_{x,y}^*\lesssim 0.3$ and out-of-plane $g$-factor $g_z^*\gtrsim 10$. As a consequence, Rabi hot spots and dephasing sweet lines are extremely sharp and call for a careful alignment of the magnetic field in Ge spin qubit devices. We investigate how the $g$-factors can be engineered by strains. We show that uniaxial strains can raise in-plane $g$-factors above unity while leaving $g_z^*$ essentially constant. We discuss how the etching of an elongated mesa in a strained buffer can actually induce uniaxial (but inhomogeneous) strains in the heterostructure. This broadens the operational magnetic field range and enables spin manipulation by shuttling holes between neighboring dots with different $g$-factors.

cond-mat.mes-hall↗

arXiv2024

Geometry of the dephasing sweet spots of spin-orbit qubits

The dephasing time of spin-orbit qubits is limited by the coupling with electrical and charge noise. However, there may exist "dephasing sweet spots" where the qubit decouples (to first order) from the noise so that the dephasing time reaches a maximum. Here we discuss the nature of the dephasing sweet spots of a spin-orbit qubit electrically coupled to some fluctuator. We characterize the Zeeman energy $E_\mathrm{Z}$ of this qubit by the tensor $G$ such that $E_\mathrm{Z}=μ_B\sqrt{\vec{B}^\mathrm{T}G\vec{B}}$ (with $μ_B$ the Bohr magneton and $\vec{B}$ the magnetic field), and its response to the fluctuator by the derivative $G^\prime$ of $G$ with respect to the fluctuating field. The geometrical nature of the sweet spots on the unit sphere describing the magnetic field orientation depends on the sign of the eigenvalues of $G^\prime$. We show that sweet spots usually draw lines on this sphere. We then discuss how to characterize the electrical susceptibility of a spin-orbit qubit with test modulations on the gates. We apply these considerations to a Ge/GeSi spin qubit heterostructure, and discuss the prospects for the engineering of sweet spots.

cond-mat.mes-hall↗

arXiv2019

Deep execution monitor for robot assistive tasks

We consider a novel approach to high-level robot task execution for a robot assistive task. In this work we explore the problem of learning to predict the next subtask by introducing a deep model for both sequencing goals and for visually evaluating the state of a task. We show that deep learning for monitoring robot tasks execution very well supports the interconnection between task-level planning and robot operations. These solutions can also cope with the natural non-determinism of the execution monitor. We show that a deep execution monitor leverages robot performance. We measure the improvement taking into account some robot helping tasks performed at a warehouse.

cs.AI↗

arXiv2019

Visual search and recognition for robot task execution and monitoring

Visual search of relevant targets in the environment is a crucial robot skill. We propose a preliminary framework for the execution monitor of a robot task, taking care of the robot attitude to visually searching the environment for targets involved in the task. Visual search is also relevant to recover from a failure. The framework exploits deep reinforcement learning to acquire a "common sense" scene structure and it takes advantage of a deep convolutional network to detect objects and relevant relations holding between them. The framework builds on these methods to introduce a vision-based execution monitoring, which uses classical planning as a backbone for task execution. Experiments show that with the proposed vision-based execution monitor the robot can complete simple tasks and can recover from failures in autonomy.

cs.AI↗

arXiv2019

Anticipation and next action forecasting in video: an end-to-end model with memory

Action anticipation and forecasting in videos do not require a hat-trick, as far as there are signs in the context to foresee how actions are going to be deployed. Capturing these signs is hard because the context includes the past. We propose an end-to-end network for action anticipation and forecasting with memory, to both anticipate the current action and foresee the next one. Experiments on action sequence datasets show excellent results indicating that training on histories with a dynamic memory can significantly improve forecasting performance.

cs.CV↗