arXiv subjects

Márton Trencséni

Publications and source records attributed to Márton Trencséni.

11 recordsLinked to original sources

arXiv2026

The Unreasonable Effectiveness of Monte Carlo Simulations in A/B Testing

This paper examines the use of Monte Carlo simulations to understand statistical concepts in A/B testing and Randomized Controlled Trials (RCTs). We discuss the applicability of simulations in understanding false positive rates and estimating statistical power, implementing variance reduction techniques and examining the effects of early stopping. By comparing frequentist and Bayesian approaches, we illustrate how simulations can clarify the relationship between p-values and posterior probabilities, and the validity of such approximations. The study also references how Monte Carlo simulations can be used to understand network effects in RCTs on social networks. Our findings show that Monte Carlo simulations are an effective tool for experimenters to deepen their understanding and ensure their results are statistically valid and practically meaningful.

stat.AP↗

arXiv2026

PaxosLease in Relativistic Inertial Frames

A lease grants one holder exclusive authority over a resource for a fixed term, and its safety property guarantees that at most one holder has the lease at any time. Under special relativity ``at any time'' has no frame-independent meaning for participants in relative motion. This note first restates the classical safety property with light cones, requiring a holder's expiry to lie in the causal past of any later acquisition by a different holder, an order every observer agrees about. Second, the paper modifies classical PaxosLease: the core change is that the containment and quarantine rules each gain a factor of $k = \sqrt{(1+β)/(1-β)}$, the relativistic Doppler factor, where $β$ is relative speed as a fraction of the speed of light.

cs.DC↗

arXiv2026

PaxosLease Revisited: A Checked Model of Diskless Distributed Leases

PaxosLease is a protocol by which a quorum of acceptors grants time-bounded exclusive ownership with no durable acceptor lease state and no disk write on the lease acquisition path. This paper gives a precise, machine-checked statement of the protocol and of its standard use, electing a Multi-Paxos leader. Formalizing and model checking the original protocol changes three rules of its acceptor: two are required for safety, the third allows shorter restart quarantines. The protocol is formalized in TLA+ and checked by TLC, its timing arithmetic is proved in TLAPS, and an executable Python model demonstrates the distributed algorithm for human readers.

cs.DC↗

arXiv2023

Monte Carlo Experiments of Network Effects in Randomized Controlled Trials

I run Monte Carlo simulations of content production over random Watts-Strogatz graphs to show various effects relevant to modeling and understanding Randomized Controlled Trials on social networks: the network effect, spillover effect, experiment dampening effect, intrinsic dampening effect, clustering effect, degree distribution effect and the experiment size effect. I will also define some simple metrics to measure their strength. When running experiments these potentially unexpected effects must be understood and controlled for in some manner, such as modeling the underlying graph structure to establish a baseline.

cs.SI↗

arXiv2013

Photo-Met: a non-parametric method for estimating stellar metallicity from photometric observations

Getting spectra at good signal-to-noise ratios takes orders of magnitudes more time than photometric observations. Building on the technique developed for photometric redshift estimation of galaxies, we develop and demonstrate a non-parametric photometric method for estimating the chemical composition of galactic stars. We investigate the efficiency of our method using spectroscopically determined stellar metallicities from SDSS DR7. The technique is generic in the sense that it is not restricted to certain stellar types or stellar parameter ranges and makes it possible to obtain metallicities and error estimates for a much larger sample than spectroscopic surveys would allow. We find that our method performs well, especially for brighter stars and higher metallicities and, in contrast to many other techniques, we are able to reliably estimate the error of the predicted metallicities.

astro-ph.IM↗

arXiv2013

ScalienDB: Designing and Implementing a Distributed Database using Paxos

ScalienDB is a scalable, replicated database built on top of the Paxos algorithm. It was developed from 2010 to 2012, when the startup backing it failed. This paper discusses the design decisions of the distributed database, describes interesting parts of the C++ codebase and enumerates lessons learned putting ScalienDB into production at a handful of clients. The source code is available on Github under the AGPL license, but it is no longer developed or maintained.

cs.DB↗

arXiv2012

Plane-Sweep Incremental Algorithm: Computing Delaunay Tessellations of Large Datasets

We present the plane-sweep incremental algorithm, a hybrid approach for computing Delaunay tessellations of large point sets whose size exceeds the computer's main memory. This approach unites the simplicity of the incremental algorithms with the comparatively low memory requirements of plane-sweep approaches. The procedure is to first sort the point set along the first principal component and then to sequentially insert the points into the tessellation, essentially simulating a sweeping plane. The part of the tessellation that has been passed by the sweeping plane can be evicted from memory and written to disk, limiting the memory requirement of the program to the "thickness" of the data set along its first principal component. We implemented the algorithm and used it to compute the Delaunay tessellation and Voronoi partition of the Sloan Digital Sky Survey magnitude space consisting of 287 million points.

cs.CG↗

arXiv2012

Spatial Indexing of Large Multidimensional Databases

Scientific endeavors such as large astronomical surveys generate databases on the terabyte scale. These, usually multidimensional databases must be visualized and mined in order to find interesting objects or to extract meaningful and qualitatively new relationships. Many statistical algorithms required for these tasks run reasonably fast when operating on small sets of in-memory data, but take noticeable performance hits when operating on large databases that do not fit into memory. We utilize new software technologies to develop and evaluate fast multidimensional indexing schemes that inherently follow the underlying, highly non-uniform distribution of the data: they are layered uniform grid indices, hierarchical binary space partitioning, and sampled flat Voronoi tessellation of the data. Our working database is the 5-dimensional magnitude space of the Sloan Digital Sky Survey with more than 270 million data points, where we show that these techniques can dramatically speed up data mining operations such as finding similar objects by example, classifying objects or comparing extensive simulation sets with observations. We are also developing tools to interact with the multidimensional database and visualize the data at multiple resolutions in an adaptive manner.

cs.DB↗

arXiv2012

PaxosLease: Diskless Paxos for Leases

This paper describes PaxosLease, a distributed algorithm for lease negotiation. PaxosLease is based on Paxos, but does not require disk writes or clock synchrony. PaxosLease is used for master lease negotation in the open-source Keyspace and ScalienDB replicated key-value stores.

cs.DC↗

arXiv2012

Keyspace: A Consistently Replicated, Highly-Available Key-Value Store

This paper describes the design and architecture of Keyspace, a distributed key-value store offering strong consistency, fault-tolerance and high availability. The source code is available under the open-source AGPL license for Linux, Windows and BSD-like platforms. As of 2012, Keyspace is no longer undergoing active development.

cs.DB↗

arXiv2012

Pure Lattice Gauge Theory in the Expanding Universe

Using Monte Carlo methods, I study the thermodynamic properties of Z_2 Abelian lattice gauge theory in flat, homogeneous, isotropic, expanding spacetimes characterized by the scale factor a(t). The presence of the scale factor introduces a spacelike domain wall inside the four dimensional spacetime lattice where the phase transition associated with deconfinement occurs. The resulting theory appears identical to introducing a time-dependent effective coupling beta_eff, and could serve as a simple model for the electroweak phase transition of the early Universe.

hep-lat↗