arXiv subjects

Mohammad Rashid

Publications and source records attributed to Mohammad Rashid.

6 recordsLinked to original sources

arXiv2026

Adaptive Ad Load Design for Sponsored Search Markets: Evidence, Theory, and Deployment

Ad-load design is a central supply-side decision in sponsored search: more sponsored slots can raise revenue, but may crowd out organic results and degrade user outcomes. We study this trade-off using a large-scale randomized field experiment on an Android app store, where over five million users are exposed to one through six sponsored slots. Increasing ad load raises revenue by up to 43%, but reduces total search conversions by up to 5% and daily engagement by up to 2.2%. These average effects mask substantial heterogeneity: additional slots generate large revenue gains for high-ad-conversion queries, but little or negative marginal revenue for low-conversion queries. The trade-off also shifts within query as advertiser composition changes, such as brand-advertiser presence. Motivated by these findings, we design and deploy a novel adaptive algorithm -- exploration-augmented Locally Adaptive Ad Load (e-LAAL). e-LAAL combines LAAL, a model-free query-level decision rule that updates ad-load recommendations using recent outcomes, with static exploration arms that maintain support and provide fixed-policy counterfactual benchmarks. We provide a finite-time dynamic-regret guarantee for the e-LAAL architecture. In a platform-level production deployment serving 22.3 million users and 77.6 million searches, e-LAAL improves the empirical revenue--conversion trade-off relative to deployed static benchmarks and outperforms uniform and historical query-dependent static benchmarks.

cs.LG↗

arXiv2025

Passive Microwave Tag Classification Using RF Fingerprinting and Machine Learning

We present an approach to identifying wireless microwave tags using radio frequency (RF) fingerprinting and machine learning. The tags are designed for low cost and simplicity, consisting of only two antennas and a single nonlinear element (a diode). An interrogating transceiver transmits a signal consisting of a set of individual frequency tones that is captured by the tag. The signal response of the diode is nonlinear, and can be represented by an infinite power series, the coefficients of which are similar but not identical for different physical diodes due to small manufacturing perturbations. The small differences in the signal responses manifest in the spectral signal response of the tag, which is retransmitted back to the interrogating transceiver. Input into machine learning algorithms, the slight differences in the spectral responses of the diodes can be used to uniquely identify devices. To demonstrate the concept, we designed 2.0 GHz tags consisting of patch antennas and a single diode, along with a bi-static radar system operating at the 2.0 GHz 802.11 Wi-Fi band transmitting multi-tone continuous wave signals representing common 802.11 training fields. The received signals were processed using a set of algorithms for comparison purposes. A real-time classification accuracy of 95% between two tags was achieved.

eess.SP↗

arXiv2025

Auctions Meet Bandits: An Empirical Analysis

Sponsored search positions are typically allocated through real-time auctions, where the outcomes depend on advertisers' quality-adjusted bids - the product of their bids and quality scores. Although quality scoring helps promote ads with higher conversion outcomes, setting these scores for new advertisers in any given market is challenging, leading to the cold-start problem. To address this, platforms incorporate multi-armed bandit algorithms in auctions to balance exploration and exploitation. However, little is known about the optimal exploration strategies in such auction environments. We utilize data from a leading Asian mobile app store that places sponsored ads for keywords. The platform employs a Thompson Sampling algorithm within a second-price auction to learn quality scores and allocate a single sponsored position for each keyword. We empirically quantify the gains from optimizing exploration under this combined auction-bandit model and show that this problem differs substantially from the canonical bandit problem. Drawing on these empirical insights, we propose a customized exploration strategy in which the platform adjusts the exploration levels for each keyword according to its characteristics. We derive the Pareto frontier for revenue and efficiency and provide actionable policies, demonstrating substantial gains for the platform on both metrics when using a tailored exploration approach.

cs.GT↗

arXiv2022

Entropic Weighted Rank Function

It is known that the entropy function over a set of jointly distributed random variables is a submodular set function. However, not any submodular function is of this form. In this paper, we consider a family of submodular set functions, called weighted rank functions of matroids, and study the necessary or sufficient conditions under which they are entropic. We prove that weighted rank functions are located on the boundary of the submodularity cone. For the representable matroids over a characteristic 2 field, we show that the integer valued weighted rank functions are entropic. We derive a necessary condition for constant weight rank functions to be entropic and show that for the case of graphic matroids, this condition is indeed sufficient. Since these functions generalize the rank of a matroid, our findings generalize some of the results of Abbe et. al. about entropic properties of the rank function of matroids.

cs.IT↗

arXiv2021

Multi-Feasibility Variable Selection

This paper is the report of the problem proposed for the !Optimizer 2021 competition, and the solutions of the gold medalist team, i.e., the Panda team. The competition was held in two stages, the research and development stage and a two-week contest stage, consisting of five rounds, and seven teams succeeded in finishing both stages to the end. In this joint report of the winner team Panda and the problem design committee coordinated by Mojtaba Tefagh, we first explain each of the five rounds and then provide the solutions proposed by our team (Panda) to fulfill the required tasks in the fastest and most accurate way. Afterward, some preprocessing and data manipulating ideas used to enhance the algorithms would be presented. All codes are written in the Julia language, which showed a better performance than Python on optimization problems in our comparisons during the R&D stage, and are publicly available in the Github repository: https://github.com/Optimizer-Competition-Panda

cs.DC↗

arXiv2018

Fibonacci self-reciprocal polynomials and Fibonacci permutation polynomials

Let $p$ be a prime. In this paper, we give a complete classification of self-reciprocal polynomials arising from Fibonacci polynomials over $\mathbb{Z}$ and $\mathbb{Z}_p$, where $p=2$ and $p>5$. We also present some partial results when $p=3, 5$. We also compute the first and second moments of Fibonacci polynomials $f_{n}(x)$ over finite fields, which give necessary conditions for Fibonacci polynomials to be permutation polynomials over finite fields.

math.NT↗