arXiv subjects

Nihar Shah

Publications and source records attributed to Nihar Shah.

8 recordsLinked to original sources

arXiv2026

Distributed Training using an Intelligent Network

Distributed training across a wide area network (WAN) is challenging, as continuous parameter exchange by islands of compute is constrained by limited bandwidth, high latency, and uneven topology. We propose making the network an active participant in training. On the systems side, such networks should leverage (i) multicast technology to replicate outbound traffic and (ii) in-line FPGAs to aggregate inbound traffic, to ease egress and ingress bottlenecks. These technologies are used for training across workers within a data center, but this paper extends them to the WAN. On the algorithms side, we develop an optimization framework that produces rich synchronization schedules (namely, rotating cliques of islands) around the underlying network topology and these technologies, to maximize information exchange. Finally, we illustrate this on a nine-city topology modeled on the DoubleZero network, a live programmable WAN equipped with both technologies, and show how the optimal schedules shift with the network's capabilities. Together, these can narrow the gap to the gold standard of colocated training.

cs.LG↗

arXiv2025

Benchmarking Fraud Detectors on Private Graph Data

We introduce the novel problem of benchmarking fraud detectors on private graph-structured data. Currently, many types of fraud are managed in part by automated detection algorithms that operate over graphs. We consider the scenario where a data holder wishes to outsource development of fraud detectors to third parties (e.g., vendors or researchers). The third parties submit their fraud detectors to the data holder, who evaluates these algorithms on a private dataset and then publicly communicates the results. We propose a realistic privacy attack on this system that allows an adversary to de-anonymize individuals' data based only on the evaluation results. In simulations of a privacy-sensitive benchmark for facial recognition algorithms by the National Institute of Standards and Technology (NIST), our attack achieves near perfect accuracy in identifying whether individuals' data is present in a private dataset, with a True Positive Rate of 0.98 at a False Positive Rate of 0.00. We then study how to benchmark algorithms while satisfying a formal differential privacy (DP) guarantee. We empirically evaluate two classes of solutions: subsample-and-aggregate and DP synthetic graph data. We demonstrate through extensive experiments that current approaches do not provide utility when guaranteeing DP. Our results indicate that the error arising from DP trades off between bias from distorting graph structure and variance from adding random noise. Current methods lie on different points along this bias-variance trade-off, but more complex methods tend to require high-variance noise addition, undermining utility.

cs.CR↗

arXiv2024

Causal Effect of Group Diversity on Redundancy and Coverage in Peer-Reviewing

A large host of scientific journals and conferences solicit peer reviews from multiple reviewers for the same submission, aiming to gather a broader range of perspectives and mitigate individual biases. In this work, we reflect on the role of diversity in the slate of reviewers assigned to evaluate a submitted paper as a factor in diversifying perspectives and improving the utility of the peer-review process. We propose two measures for assessing review utility: review coverage -- reviews should cover most contents of the paper -- and review redundancy -- reviews should add information not already present in other reviews. We hypothesize that reviews from diverse reviewers will exhibit high coverage and low redundancy. We conduct a causal study of different measures of reviewer diversity on review coverage and redundancy using observational data from a peer-reviewed conference with approximately 5,000 submitted papers. Our study reveals disparate effects of different diversity measures on review coverage and redundancy. Our study finds that assigning a group of reviewers that are topically diverse, have different seniority levels, or have distinct publication networks leads to broader coverage of the paper or review criteria, but we find no evidence of an increase in coverage for reviewer slates with reviewers from diverse organizations or geographical locations. Reviewers from different organizations, seniority levels, topics, or publications networks (all except geographical diversity) lead to a decrease in redundancy in reviews. Furthermore, publication network-based diversity alone also helps bring in varying perspectives (that is, low redundancy), even within specific review criteria. Our study adopts a group decision-making perspective for reviewer assignments in peer review and suggests dimensions of diversity that can help guide the reviewer assignment process.

cs.DL↗

arXiv2024

Weighted Regression with Sybil Networks

In many online domains, Sybil networks -- or cases where a single user assumes multiple identities -- is a pervasive feature. This complicates experiments, as off-the-shelf regression estimators at least assume known network topologies (if not fully independent observations) when Sybil network topologies in practice are often unknown. The literature has exclusively focused on techniques to detect Sybil networks, leading many experimenters to subsequently exclude suspected networks entirely before estimating treatment effects. I present a more efficient solution in the presence of these suspected Sybil networks: a weighted regression framework that applies weights based on the probabilities that sets of observations are controlled by single actors. I show in the paper that the MSE-minimizing solution is to set the weight matrix equal to the inverse of the expected network topology. I demonstrate the methodology on simulated data, and then I apply the technique to a competition with suspected Sybil networks run on the Sui blockchain and show reductions in the standard error of the estimate by 6 - 24%.

stat.ME↗

arXiv2023

Triangle Fees

Triangle fees are a novel fee structure for AMMs, in which marginal fees are decreasing in a trade's size. That decline is proportional to the movement in the AMM's implied price, i.e. for every basis point the trade moves the ratio of assets, the marginal fee declines by a basis point. These fees create incentives that protect against price staleness, while still allowing the AMM to earn meaningful fee revenue. Triangle fees can strictly improve the Pareto frontier of price accuracy versus losses generated by the status quo of constant fee mechanisms.

q-fin.MF↗

arXiv2023

A Short Note on Setting Swap Parameters

This short note illustrates the theoretical solution to a trader determining how to optimally swap her wealth into a target asset through on-chain operations. It offers the framework to solve optimal slippage parameters and optimal trade size.

q-fin.MF↗

arXiv2023

Time is Money: Strategic Timing Games in Proof-of-Stake Protocols

We propose a model suggesting that honest-but-rational consensus participants may play timing games, and strategically delay their block proposal to optimize MEV capture, while still ensuring the proposal's timely inclusion in the canonical chain. In this context, ensuring economic fairness among consensus participants is critical to preserving decentralization. We contend that a model grounded in honest-but-rational consensus participation provides a more accurate portrayal of behavior in economically incentivized systems such as blockchain protocols. We empirically investigate timing games on the Ethereum network and demonstrate that while timing games are worth playing, they are not currently being exploited by consensus participants. By quantifying the marginal value of time, we uncover strong evidence pointing towards their future potential, despite the limited exploitation of MEV capture observed at present.

cs.GT↗

arXiv2022

The Square Root Agreement Rule for Incentivizing Truthful Feedback on Online Platforms

A major challenge in obtaining evaluations of products or services on e-commerce platforms is eliciting informative responses in the absence of verifiability. This paper proposes the Square Root Agreement Rule (SRA): a simple reward mechanism that incentivizes truthful responses to objective evaluations on such platforms. In this mechanism, an agent gets a reward for an evaluation only if her answer matches that of her peer, where this reward is inversely proportional to a popularity index of the answer. This index is defined to be the square root of the empirical frequency at which any two agents performing the same evaluation agree on the particular answer across evaluations of similar entities operating on the platform. Rarely agreed-upon answers thus earn a higher reward than answers for which agreements are relatively more common. We show that in the many tasks regime, the truthful equilibrium under SRA is strictly payoff-dominant across large classes of natural equilibria that could arise in these settings, thus increasing the likelihood of its adoption. While there exist other mechanisms achieving such guarantees, they either impose additional assumptions on the response distribution that are not generally satisfied for objective evaluations or they incentivize truthful behavior only if each agent performs a prohibitively large number of evaluations and commits to using the same strategy for each evaluation. SRA is the first known incentive mechanism satisfying such guarantees without imposing any such requirements. Moreover, our empirical findings demonstrate the robustness of the incentive properties of SRA in the presence of mild subjectivity or observational biases in the responses. These properties make SRA uniquely attractive for administering reward-based incentive schemes (e.g., rebates, discounts, reputation scores, etc.) on online platforms.

cs.GT↗