arXiv subjects

Akshit Kumar

Publications and source records attributed to Akshit Kumar.

7 recordsLinked to original sources

arXiv2025

Digital Twins as Funhouse Mirrors: Five Key Distortions

Scientists and practitioners are increasingly moving to deploy digital twins--LLM-based models of real individuals--across social science and policy research. We conduct 19 pre-registered studies spanning 164 diverse outcomes (e.g., attitudes toward hiring algorithms, intentions to share misinformation), comparing human responses to those of their corresponding digital twins, which are trained on each individual's prior responses to over 500 questions. We establish an empirical benchmark for digital twin performance: their predictions are only modestly more accurate than those of a homogeneous base LLM and exhibit weak correlation with human responses (average $r = 0.20$). To inform future development, we identify five systematic distortions in digital twin behavior: (i) insufficient individuation, (ii) stereotyping, (iii) representation bias, (iv) ideological bias, and (v) hyper-rationality. Finally, we release our full dataset and code as a standardized testbed for evaluating and improving digital twin methodologies. Together, our findings caution against premature deployment while laying the groundwork for a transparent, replicable, and iterative science of responsible digital twin development.

cs.CY↗

arXiv2025

What Is Your AI Agent Buying? Evaluation, Biases, Model Dependence, & Emerging Implications for Agentic E-Commerce

Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, AI agents can parse webpages or leverage APIs to view, evaluate and choose products. We investigate the behavior of AI agents using ACES, a provider-agnostic framework for auditing agent decision-making. We reveal that agents can exhibit choice homogeneity, often concentrating demand on a few ``modal'' products while ignoring others entirely. Yet, these preferences are unstable: model updates can drastically reshuffle market shares. Furthermore, randomized trials show that while agents have improved over time on simple tasks with a clearly identified best choice, they exhibit strong position biases -- varying across providers and model versions, and persisting even in text-only "headless" interfaces -- undermining any universal notion of a ``top'' rank. Agents also consistently penalize sponsored tags while rewarding platform endorsements, and sensitivities to price, ratings, and reviews vary sharply across models. Finally, we demonstrate that sellers can respond: a seller-side agent making simple, query-conditional description tweaks can drive significant gains in market share. These findings reveal that agentic markets are volatile and fundamentally different from human-centric commerce, highlighting the need for continuous auditing and raising questions for platform design, seller strategy and regulation.

cs.AI↗

arXiv2025

Performance of LLMs on Stochastic Modeling Operations Research Problems: From Theory to Practice

Large language models (LLMs) have exhibited expert-level capabilities across various domains. However, their abilities to solve problems in Operations Research (OR) -- the analysis and optimization of mathematical models derived from real-world problems or their verbal descriptions -- remain underexplored. In this work, we take a first step toward evaluating LLMs' abilities to solve stochastic modeling problems, a core class of OR problems characterized by uncertainty and typically involving tools from probability, statistics, and stochastic processes. We manually procure a representative set of graduate-level homework and doctoral qualification-exam problems and test LLMs' abilities to solve them. We further leverage SimOpt, an open-source library of simulation-optimization problems and solvers, to investigate LLMs' abilities to make real-world decisions under uncertainty. Our results show that, though a nontrivial amount of work is still needed to reliably automate the stochastic modeling pipeline in reality, state-of-the-art LLMs demonstrate proficiency on par with human experts in both classroom and practical settings. These findings highlight the potential of building AI agents that assist OR researchers and amplify the real-world impact of OR through automation.

cs.AI↗

arXiv2025

Impact of Rankings and Personalized Recommendations in Marketplaces

Decision-making often requires individuals to navigate large sets of options with incomplete information and imperfectly formed preferences. Information provisioning tools, such as public rankings and personalized recommendations, have become central to guiding these choices, yet their welfare implications across different market environments remain poorly understood. This paper studies a stylized large-market model to quantify the aggregate value of these tools under uncapacitated supply and capacitated supply. Agent utility is a weighted combination of a common term (population-level quality) and an idiosyncratic term (individual-specific fit). Agents observe noisy signals of these components: public rankings improve estimates of common quality, while personalized recommendations additionally reveal idiosyncratic fit. In uncapacitated settings, both tools improve welfare through better selection. Their relative value is governed by preference heterogeneity: rankings are highly effective when preferences are relatively homogeneous, whereas the marginal value of personalization grows as heterogeneity increases. In stark contrast, under capacity constraints, a strict conservation logic applies: learning common quality primarily reshuffles assignments without increasing aggregate average agent welfare. However, revealing idiosyncratic fit generates substantial welfare gains by mitigating congestion on overdemanded items and unlocking match-specific value. We demonstrate that these core insights are robust across model extensions including alternative extreme-value tail distributions and heterogeneous supply-side priorities. Finally, introducing strategic supply-side pricing reveals a stark distributional dichotomy: while information expands total surplus, the newly generated value is ultimately extracted by the short side of the market.

econ.TH↗

arXiv2024

The Fault in Our Recommendations: On the Perils of Optimizing the Measurable

Recommendation systems are widespread, and through customized recommendations, promise to match users with options they will like. To that end, data on engagement is collected and used. Most recommendation systems are ranking-based, where they rank and recommend items based on their predicted engagement. However, the engagement signals are often only a crude proxy for utility, as data on the latter is rarely collected or available. This paper explores the following question: By optimizing for measurable proxies, are recommendation systems at risk of significantly under-delivering on utility? If so, how can one improve utility which is seldom measured? To study these questions, we introduce a model of repeated user consumption in which, at each interaction, users select between an outside option and the best option from a recommendation set. Our model accounts for user heterogeneity, with the majority preferring ``popular'' content, and a minority favoring ``niche'' content. The system initially lacks knowledge of individual user preferences but can learn them through observations of users' choices over time. Our theoretical and numerical analysis demonstrate that optimizing for engagement can lead to significant utility losses. Instead, we propose a utility-aware policy that initially recommends a mix of popular and niche content. As the platform becomes more forward-looking, our utility-aware policy achieves the best of both worlds: near-optimal utility and near-optimal engagement simultaneously. Our study elucidates an important feature of recommendation systems; given the ability to suggest multiple items, one can perform significant exploration without incurring significant reductions in engagement. By recommending high-risk, high-reward items alongside popular items, systems can enhance discovery of high utility items without significantly affecting engagement.

cs.IR↗

arXiv2022

Dynamic Resource Allocation: Algorithmic Design Principles and Spectrum of Achievable Performances

Dynamic resource allocation problems are ubiquitous, arising in inventory management, order fulfillment, online advertising, and other applications. We initially focus on one of the simplest models of online resource allocation: the multisecretary problem. In the multisecretary problem, a decision maker sequentially hires up to $B$ out of $T$ candidates, and candidate ability values are drawn i.i.d. from a distribution $F$ on $[0,1]$. First, we investigate fundamental limits on performance as a function of the value distribution under consideration. We quantify performance in terms of regret, defined as the additive loss relative to the best performance achievable in hindsight. We present a novel fundamental regret lower bound scaling of $\Omega(T^{1/2 - 1/2(1 + \beta)})$ for distributions with gaps in their support, with $\beta$ quantifying the mass accumulation of types (values) around these gaps. This lower bound contrasts with the constant and logarithmic regret guarantees shown to be achievable in prior work, under specific assumptions on the value distribution. Second, we introduce a novel algorithmic principle, Conservativeness with respect to Gaps (CwG), which yields near-optimal performance with regret scaling of $\tilde{O}(T^{1/2 - 1/2(1 + \beta)})$ for any distribution in a class parameterized by the mass accumulation parameter $\beta$. We then turn to operationalizing the CwG principle across dynamic resource allocation problems. We study a general and practical algorithm, Repeatedly Act using Multiple Simulations (RAMS), which simulates possible futures to estimate a hindsight-based approximation of the value-to-go function. We establish that this algorithm inherits theoretical performance guarantees of algorithms tailored to the distribution of resource requests, including our CwG-based algorithm, and find that it outperforms them in numerical experiments.

math.PR↗

arXiv2020

Breaking the Unit Throughput Barrier in Distributed Systems

A multi-level random power transmit strategy that is used in conjunction with a random access protocol (RAP) (e.g. ALOHA, IRSA) is proposed to fundamentally increase the throughput in a distributed communication network. A SIR model is considered, where a packet is decodable as long as its SIR is above a certain threshold. In a slot chosen for transmission by a RAP, a packet is transmitted with power level chosen according to a distribution, such that multiple packets sent by different nodes can be decoded at the receiver in a single slot, by ensuring that their SIRs are above the threshold with successive interference cancelation. Since the network is distributed this is a challenging task, and we provide structural results that aid in finding the achievable throughputs, together with upper bounds on the maximum throughput possible. The achievable throughput and the upper bounds are shown to be close with the help of comprehensive simulations. The main takeaway is that the throughput of more than 1 is possible in a distributed network, by using a judicious choice of power level distribution in conjuction with a RAP.

cs.IT↗