arXiv subjects

Sayan Patra

Publications and source records attributed to Sayan Patra.

8 recordsLinked to original sources

arXiv2026

Design and fabrication of a micro-ion trap with a 3D-printed loading zone for improved hot-ion capture

We leverage recent advances in 3D-printing technology to design and fabricate a micro-ion trap with a spatially distinct loading zone for more efficient loading of ions from effusive thermal ovens. The design reduces the Mathieu-$q$ parameter in the loading zone by increasing the ion-electrode separation $r_0$, thereby potentially facilitating more effective laser cooling of hot ions. This circumvents the temporary thermal instability that arises when the rf potential is reduced during ion loading, a common practice to enable efficient laser cooling of hot ions. Simulations predict that expanding $r_0$ maintains a high trapped ion fraction from a simulated thermal source across a wide range of Mathieu-$q$ parameters. We demonstrate the manufacturability of this design by 3D-printing the rf rails of a four-rod ion trap and discuss the limitations imposed by state-of-the-art additive manufacturing techniques. We briefly compare hot-ion capture in the three-dimensional design presented here with that in a representative planar trap, illustrating one instance in which the former may be better for loading. The article concludes with an outlook for how this design may be incorporated into a quantum-CCD architecture to enhance ion loading and reduce associated experimental overheads.

physics.atom-ph↗

arXiv2026

Endcap-Type Paul Trap for Precision Spectroscopy and Studies of Controlled Interactions

We present the design and fabrication of an endcap-type Paul trap. The trap is designed for studies with Ca$^{+}$ and Yb$^{+}$. The design, fabrication process, and characterization are presented in detail with a focus on trapping a single compensated ion at the rf node. A custom-built imaging system of $NA = 0.14$ and magnification $\approx 22 \times$ performs close to diffraction-limit and resolves multi-ion clusters. Controlled ion loading and characterization of the trap are performed using $^{40}$Ca$^{+}$. The experimentally determined quadrupole coefficient of the trap is $\approx 0.3$, which is very close to the design value. The relative frequency shift along the spectroscopy beam due to excess micromotion (EMM) is at the level of $3.5\times 10^{-18}$ for $^{40}$Ca$^{+}$. Applications of this trap encompass single-ion-based optical frequency standards, tests of fundamental physics, the study of mesoscopic Coulomb clusters, and the controlled interaction of a single ion with co-trapped atoms.

physics.atom-ph↗

arXiv2022

Greykite: Deploying Flexible Forecasting at Scale at LinkedIn

Forecasts help businesses allocate resources and achieve objectives. At LinkedIn, product owners use forecasts to set business targets, track outlook, and monitor health. Engineers use forecasts to efficiently provision hardware. Developing a forecasting solution to meet these needs requires accurate and interpretable forecasts on diverse time series with sub-hourly to quarterly frequencies. We present Greykite, an open-source Python library for forecasting that has been deployed on over twenty use cases at LinkedIn. Its flagship algorithm, Silverkite, provides interpretable, fast, and highly flexible univariate forecasts that capture effects such as time-varying growth and seasonality, autocorrelation, holidays, and regressors. The library enables self-serve accuracy and trust by facilitating data exploration, model configuration, execution, and interpretation. Our benchmark results show excellent out-of-the-box speed and accuracy on datasets from a variety of domains. Over the past two years, Greykite forecasts have been trusted by Finance, Engineering, and Product teams for resource planning and allocation, target setting and progress tracking, anomaly detection and root cause analysis. We expect Greykite to be useful to forecast practitioners with similar applications who need accurate, interpretable forecasts that capture complex dynamics common to time series related to human activity.

cs.LG↗

arXiv2022

Proton-Electron Mass Ratio from Laser Spectroscopy of HD$^+$ at the Part-Per-Trillion Level

Accepted values of the masses of several subatomic particles have been under debate since recent measurements in Penning traps produced more precise yet incompatible results, implying possible inconsistencies in closely related physical constants like the proton-electron and deuteron-proton mass ratios. These quantities also influence the predicted vibrational spectrum of the deuterated molecular hydrogen ion in its electronic ground state, of which we measured the v=0 - 9 overtone transition frequency with an uncertainty of 2.9 parts-per-trillion through Doppler-free two-photon laser spectroscopy. Leveraging high-precision ab initio calculations we convert our measurement to tight constraints on the proton-electron and deuteron-proton mass ratios, consistent with the most recent Penning-trap determinations of these quantities, and yielding a new value of the proton-electron mass ratio with an unprecedented precision of 21 parts-per-trillion.

physics.atom-ph↗

arXiv2021

A flexible forecasting model for production systems

This paper discusses desirable properties of forecasting models in production systems. It then develops a family of models which are designed to satisfy these properties: highly customizable to capture complex patterns; accommodates a large variety of objectives; has interpretable components; produces robust results; has automatic changepoint detection for trend and seasonality; and runs fast -- making it a good choice for reliable and scalable production systems. The model allows for seasonality at various time scales, events/holidays, and change points in trend and seasonality. The volatility is fitted separately to maintain flexibility and speed and is allowed to be a function of specified features.

stat.ME↗

arXiv2018

Posterior Projection for Inference in Constrained Spaces

Estimation of parameters that obey specific constraints is crucial in statistics and machine learning; for example, when parameters are required to satisfy boundedness, monotonicity, or linear inequalities. Traditional approaches impose these constraints via constraint-specific transformations or sampling approaches, or by truncating the posterior distribution. Such methods often result in computational challenges, limited flexibility, and a lack of generality. We propose a generalized framework for constrained Bayesian inference by projecting the unconstrained posterior distribution into the space of the parameter constraints, providing a computationally efficient and easily implementable solution for a large class of problems. We rigorously establish the theoretical foundations of the projected posterior distribution, as well as providing asymptotic results for posterior consistency, posterior contraction, and optimal coverage properties. Our methodology is validated through both theoretical arguments and practical applications.

stat.ME↗

arXiv2017

Proton-electron mass ratio from HD$^+$ revisited

We present a new derivation of the proton-electron mass ratio from the hydrogen molecular ion, HD$^+$. The derivation entails the adjustment of the mass ratio in highly precise theory so as to reproduce accurately measured ro-vibrational frequencies. This work is motivated by recent improvements of the theory, as well as the more accurate value of the electron mass in the recently published CODATA-14 set of fundamental constants, which justifies using it as input data in the adjustment, rather than the proton mass value as done in previous works. This leads to significantly different sensitivity coefficients and, consequently, a different value and larger uncertainty margin of the proton-electron mass ratio as obtained from HD$^+$.

physics.atom-ph↗

arXiv2016

Ghost features in Doppler-broadened spectra of rovibrational transitions in trapped HD$^+$ ions

Doppler broadening plays an important role in laser rovibrational spectroscopy of trapped deuterated molecular hydrogen ions (HD$^+$), even at the millikelvin temperatures achieved through sympathetic cooling by laser-cooled beryllium ions. Recently, Biesheuvel \textit{et al.} \cite{Biesheuvel2016} presented a theoretical lineshape model for such transitions which not only considers linestrengths and Doppler broadening, but also the finite sample size and population redistribution by blackbody radiation, which are important in view of the long storage and probe times achievable in ion traps. Here, we employ the rate equation model developed by Biesheuvel \textit{et al.} to theoretically study the Doppler-broadened hyperfine structure of the $(v,L):(0,3)\rightarrow(4,2)$ rovibrational transition in HD$^+$ at 1442~nm. We observe prominent yet hitherto unrecognized ghost features in the simulated spectrum, whose positions depend on the Doppler width, transition rates, and saturation levels of the hyperfine components addressed by the laser. We explain the origin and behavior of such features, and we provide a simple quantitative guideline to assess whether ghost features may appear. As such ghost features may be common to saturated Doppler-broadened spectra of rotational and vibrational transitions in trapped ions composed of partly overlapping lines, our work illustrates the necessity to use lineshape models that take into account all the relevant physics.

physics.atom-ph↗