arXiv subjects

Xiaotong Sun

Publications and source records attributed to Xiaotong Sun.

14 recordsLinked to original sources

arXiv2026

Substitution or Complement? Uncovering the Interplay between Ride-hailing Services and Public Transit

The literature on transportation network companies (TNCs), also known as ride-hailing services, has often characterized these service providers as predominantly substitutive to public transit (PT). However, as TNC markets expand and mature, the complementary and substitutive relationships with PT may shift. To explore whether such a transformation is occurring, this study collected travel data from 96,716 ride-hailing vehicles during September 2022 in Shanghai, a city characterized by an increasingly saturated TNC market. An enhanced data-driven framework is proposed to classify TNC-PT relationships into four types: first-mile complementary, last-mile complementary, substitutive, and independent. Our findings reveal a substantial increase in the complementary ratio (9.22%) and a relative decline in the substitutive ratio (9.06%) compared to previous studies. Furthermore, to examine the nonlinear impact of various influential factors on these ratios, a machine learning method integrating categorical boosting (CatBoost) and Shapley additive explanations (SHAP) is proposed. The results show significant nonlinear effects in some variables, including the distance to the nearest metro station and the density of bus stops.

cs.SI↗

arXiv2025

Complete weight enumerators and weight hierarchies for linear codes from quadratic forms

In this paper, for an odd prime power $q$, we extend the construction of Xie et al. \cite{XOYM2023} to propose two classes of linear codes $\mathcal{C}_{Q}$ and $\mathcal{C}_{Q}'$ over the finite field $\mathbb{F}_{q}$ with at most four nonzero weights. These codes are derived from quadratic forms through a bivariate construction. We completely determine their complete weight enumerators and weight hierarchies by employing exponential sums. Most of these codes are minimal and some are optimal in the sense that they meet the Griesmer bound. Furthermore, we also establish the weight hierarchies of $\mathcal{C}_{Q,N}$ and $\mathcal{C}_{Q,N}'$, which are the descended codes of $\mathcal{C}_{Q}$ and $\mathcal{C}_{Q}'$.

cs.IT↗

arXiv2025

Cracking Instance Jigsaw Puzzles: An Alternative to Multiple Instance Learning for Whole Slide Image Analysis

While multiple instance learning (MIL) has shown to be a promising approach for histopathological whole slide image (WSI) analysis, its reliance on permutation invariance significantly limits its capacity to effectively uncover semantic correlations between instances within WSIs. Based on our empirical and theoretical investigations, we argue that approaches that are not permutation-invariant but better capture spatial correlations between instances can offer more effective solutions. In light of these findings, we propose a novel alternative to existing MIL for WSI analysis by learning to restore the order of instances from their randomly shuffled arrangement. We term this task as cracking an instance jigsaw puzzle problem, where semantic correlations between instances are uncovered. To tackle the instance jigsaw puzzles, we propose a novel Siamese network solution, which is theoretically justified by optimal transport theory. We validate the proposed method on WSI classification and survival prediction tasks, where the proposed method outperforms the recent state-of-the-art MIL competitors. The code is available at https://github.com/xiwenc1/MIL-JigsawPuzzles.

eess.IV↗

arXiv2025

On the Robotic Uncertainty of Fully Autonomous Traffic: From Stochastic Car-Following to Mobility-Safety Trade-Offs

Recent transportation research highlights the potential of autonomous vehicles (AV) to improve traffic flow mobility as they are able to maintain smaller car-following distances. However, as a unique class of ground robots, AVs are susceptible to robotic errors, particularly in their perception and control modules with imperfect sensors and actuators, leading to uncertainties in their movements and an increased risk of collisions. Consequently, conservative operational strategies, such as larger headway and slower speeds, are implemented to prioritize safety over mobility in real-world operations. To reconcile the inconsistency, this paper presents an analytical model framework that delineates the endogenous reciprocity between traffic safety and mobility that arises from AVs' robotic uncertainties. Using both realistic car-following data and a stochastic intelligent driving model (IDM), the stochastic car-following distance is derived as a key parameter, enabling analysis of single-lane capacity and collision probability. A semi-Markov process is then employed to model the dynamics of the lane capacity, and the resulting collision-inclusive capacity, representing expected lane capacity under stationary conditions, serves as the primary performance metric for fully autonomous traffic. The analytical results are further utilized to investigate the impacts of critical parameters in AV and roadway designs on traffic performance, as well as the properties of optimal speed and headway under mobility-targeted or safety-dominated management objectives. Extensions to scenarios involving multiple non-independent collisions or multi-lane traffic scenarios are also discussed, which demonstrates the robustness of the theoretical results and their practical applications.

cs.RO↗

arXiv2024

Multimodal Variational Autoencoder: a Barycentric View

Multiple signal modalities, such as vision and sounds, are naturally present in real-world phenomena. Recently, there has been growing interest in learning generative models, in particular variational autoencoder (VAE), to for multimodal representation learning especially in the case of missing modalities. The primary goal of these models is to learn a modality-invariant and modality-specific representation that characterizes information across multiple modalities. Previous attempts at multimodal VAEs approach this mainly through the lens of experts, aggregating unimodal inference distributions with a product of experts (PoE), a mixture of experts (MoE), or a combination of both. In this paper, we provide an alternative generic and theoretical formulation of multimodal VAE through the lens of barycenter. We first show that PoE and MoE are specific instances of barycenters, derived by minimizing the asymmetric weighted KL divergence to unimodal inference distributions. Our novel formulation extends these two barycenters to a more flexible choice by considering different types of divergences. In particular, we explore the Wasserstein barycenter defined by the 2-Wasserstein distance, which better preserves the geometry of unimodal distributions by capturing both modality-specific and modality-invariant representations compared to KL divergence. Empirical studies on three multimodal benchmarks demonstrated the effectiveness of the proposed method.

cs.LG↗

arXiv2024

SurvReLU: Inherently Interpretable Survival Analysis via Deep ReLU Networks

Survival analysis models time-to-event distributions with censorship. Recently, deep survival models using neural networks have dominated due to their representational power and state-of-the-art performance. However, their "black-box" nature hinders interpretability, which is crucial in real-world applications. In contrast, "white-box" tree-based survival models offer better interpretability but struggle to converge to global optima due to greedy expansion. In this paper, we bridge the gap between previous deep survival models and traditional tree-based survival models through deep rectified linear unit (ReLU) networks. We show that a deliberately constructed deep ReLU network (SurvReLU) can harness the interpretability of tree-based structures with the representational power of deep survival models. Empirical studies on both simulated and real survival benchmark datasets show the effectiveness of the proposed SurvReLU in terms of performance and interoperability. The code is available at \href{https://github.com/xs018/SurvReLU}{\color{magenta}{ https://github.com/xs018/SurvReLU}}.

cs.LG↗

arXiv2024

Planning of Truck Platooning for Road-Network Capacitated Vehicle Routing Problem

Truck platooning, a linking technology of trucks on the highway, has gained enormous attention in recent years due to its benefits in energy and operation cost savings. However, most existing studies on truck platooning limit their focus on scenarios in which each truck can serve only one customer demand and is thus with a specified origin-destination pair, so only routing and time schedules are considered. Nevertheless, in real-world logistics, each truck may need to serve multiple customers located at different places, and the operator has to determine not only the routing and time schedules of each truck but also the set of customers allocated to each truck and their sequence to visit. This is well known as a capacitated vehicle routing problem with time windows (CVRPTW), and considering the application of truck platooning in such a problem entails new modeling frameworks and tailored solution algorithms. In light of this, this study makes the first attempt to optimize the truck platooning plan for a road-network CVRPTW to minimize the total operation cost, including vehicles' fixed dispatch cost and energy cost, while fulfilling all delivery demands within their time window constraints. Specifically, the operation plan will dictate the number of trucks to be dispatched, the set of customers, and the routing and time schedules for each truck. In addition, the modeling framework is constructed based on a road network instead of a traditional customer node graph to better resemble and facilitate the platooning operation. A 3-stage algorithm embedded with a "route-then-schedule" scheme, dynamic programming, and modified insertion heuristic, is developed to solve the proposed model in a timely manner. Numerical experiments are conducted to validate the modeling framework, demonstrate the performance of the proposed solution algorithm, and quantify the benefit of truck platooning.

math.OC↗

arXiv2023

Hub-Based Platoon Formation: Optimal Release Policies and Approximate Solutions

This paper studies the optimal hub-based platoon formation at hubs along a highway under decentralized, distributed, and centralized policies. Hubs are locations along highways where trucks can wait for other trucks to form platoons. A coordinator at each hub decides the departure time of trucks, and the released trucks from the hub will form platoons. The problem is cast as an optimization problem where the objective is to maximize the platooning reward. We first show that the optimal release policy in the decentralized case, where the hubs do not exchange information, is to release all trucks at the hub when the number of trucks exceeds a threshold computed by dynamic programming. We develop efficient approximate release policies for the dependent arrival case using this result. To study the value of information exchange among hubs on platoon formation, we next study the distributed and centralized platoon formation policies which require information exchange among hubs. To this end, we develop receding horizon solutions for the distributed and centralized platoon formation at hubs using the dynamic programming technique. Finally, we perform a simulation study over three hubs in northern Sweden. The profits of the decentralized policies are shown to be approximately 3.5% lower than the distributed policy and 8% lower than the centralized release policy. This observation suggests that decentralized policies are prominent solutions for hub-based platooning as they do not require information exchange among hubs and can achieve a similar performance compared with distributed and centralized policies.

eess.SY↗

arXiv2023

NSOTree: Neural Survival Oblique Tree

Survival analysis is a statistical method employed to scrutinize the duration until a specific event of interest transpires, known as time-to-event information characterized by censorship. Recently, deep learning-based methods have dominated this field due to their representational capacity and state-of-the-art performance. However, the black-box nature of the deep neural network hinders its interpretability, which is desired in real-world survival applications but has been largely neglected by previous works. In contrast, conventional tree-based methods are advantageous with respect to interpretability, while consistently grappling with an inability to approximate the global optima due to greedy expansion. In this paper, we leverage the strengths of both neural networks and tree-based methods, capitalizing on their ability to approximate intricate functions while maintaining interpretability. To this end, we propose a Neural Survival Oblique Tree (NSOTree) for survival analysis. Specifically, the NSOTree was derived from the ReLU network and can be easily incorporated into existing survival models in a plug-and-play fashion. Evaluations on both simulated and real survival datasets demonstrated the effectiveness of the proposed method in terms of performance and interpretability.

stat.ML↗

arXiv2023

Voter Coalitions and democracy in Decentralized Finance: Evidence from MakerDAO

Decentralized Autonomous Organization (DAO) provides a decentralized governance solution through blockchain, where decision-making process relies on on-chain voting and follows majority rule. This paper focuses on MakerDAO, and we find three voter coalitions after applying clustering algorithm to voting history. The emergence of a dominant voter coalition is a signal of governance centralization in DAO, and voter coalitions have complicated influence on Maker protocol, which is governed by MakerDAO. This paper presents empirical evidence of multicoalition democracy in DAO and further contributes to the contemporary debate on whether decentralized governance is possible.

cs.SI↗

arXiv2023

Liquidity Risks in Lending Protocols: Evidence from Aave Protocol

Lending Protocols (LPs), as blockchain-based lending systems, allow any agents to borrow and lend cryptocurrencies. However, liquidity risks could occur, especially when salient loans are initiated by a particular group of borrowers. This paper proposes measurements of liquidity risks, focusing on both available liquidity and market concentration in LPs. By using Aave as a case study, we find that liquidity risks are highly volatile and show complex effects on Aave, and liquidity in Aave may affect across on-chain lending market. Compared to new users, regular users that repeatedly borrow cryptocurrencies may negatively affect Aave protocol, implying that user loyalty is a double-edged sword for LPs.

q-fin.RM↗

arXiv2023

Decentralization illusion in Decentralized Finance: Evidence from tokenized voting in MakerDAO polls

Decentralized Autonomous Organization (DAO) is very popular in Decentralized Finance (DeFi) applications as it provides a decentralized governance solution through blockchain. We analyze the governance characteristics in the Maker protocol, its stablecoin DAI and governance token Maker (MKR). To achieve that, we establish several measurements of centralized governance. Our empirical analysis investigates the effect of centralized governance over a series of factors related to MKR and DAI, such as financial, transaction, network and twitter sentiment indicators. Our results show that governance centralization influences both the Maker protocol, and the distribution of voting power matters. The main implication of this study is that centralized governance in MakerDAO very much exists, while DeFi investors face a trade-off between decentralization and performance of a DeFi protocol. This further contributes to the contemporary debate on whether DeFi can be truly decentralized. centralized governance in MakerDAO very much exists, while DeFi investors face a trade-off between efficiency and decentralization. This further contributes to the contemporary debate on whether DeFi can be truly decentralized.

cs.CR↗

arXiv2022

Bribes to Miners: Evidence from Ethereum

In blockchain, bribery is an inevitable problem since users with various goals can bribe miners by transferring cryptoassets. To alleviate the negative effects of such collusion, Ethereum blockchain implemented new transaction fee mechanism in the London Fork, which was deployed on August 5th, 2021. In this paper, we first filter potential bribery by scanning Ethereum transactions, and the potential bribers and bribees are centralized in a small group. Then we construct bribing proxies to measure the active level of bribery and then investigate the effects of bribery. Consequently, bribery can influence both Ethereum and other mainstream blockchains, in aspects of underlying cryptocurrency, transaction statistics, and network adoption. Moreover, the London Fork shows complicated effects on relationship between bribery and blockchain factors. Besides, bribery in Ethereum relates to stock markets, e.g., S&P 500 and Nasdaq, implying implicit interlinks between blockchain and traditional finance.

cs.CR↗

arXiv2022

Real-Time Cross-Fleet Pareto-Improving Truck Platoon Coordination

This paper studies a multi-fleet platoon coordination system in transport networks that deploy hubs to form trucks into platoons. The trucks belong to different fleets that are interested in increasing their profits by platooning across fleets. The profit of each fleet incorporates platooning rewards and costs for waiting at hubs. Each truck has a fixed route and a waiting time budget to spend at the hubs along its route. To ensure that all fleets are willing to participate in the system, we develop a cross-fleet Pareto-improving coordination strategy that guarantees higher fleet profits than a coordination strategy without cross-fleet platoons. By leveraging multiple hubs for platoon formation, the coordination strategy can be implemented in a real-time and distributed fashion while largely reducing the amount of travel information to be shared for system-wide coordination. We evaluate the proposed strategy in a simulation study over the Swedish transportation network. The cross-fleet platooning strategy significantly improves fleets' profits compared with single-fleet platooning, especially the profits from smaller fleets. The cross-fleet platooning strategy also shows strong competitiveness in terms of the system-wide profit compared to the case when a system planner optimizes all fleets' total profit.

eess.SY↗