arXiv subjects

Qi Yan

Publications and source records attributed to Qi Yan.

At least 55 records · Page 3Linked to original sources

arXiv2024

Harnessing the power of longitudinal medical imaging for eye disease prognosis using Transformer-based sequence modeling

Deep learning has enabled breakthroughs in automated diagnosis from medical imaging, with many successful applications in ophthalmology. However, standard medical image classification approaches only assess disease presence at the time of acquisition, neglecting the common clinical setting of longitudinal imaging. For slow, progressive eye diseases like age-related macular degeneration (AMD) and primary open-angle glaucoma (POAG), patients undergo repeated imaging over time to track disease progression and forecasting the future risk of developing disease is critical to properly plan treatment. Our proposed Longitudinal Transformer for Survival Analysis (LTSA) enables dynamic disease prognosis from longitudinal medical imaging, modeling the time to disease from sequences of fundus photography images captured over long, irregular time periods. Using longitudinal imaging data from the Age-Related Eye Disease Study (AREDS) and Ocular Hypertension Treatment Study (OHTS), LTSA significantly outperformed a single-image baseline in 19/20 head-to-head comparisons on late AMD prognosis and 18/20 comparisons on POAG prognosis. A temporal attention analysis also suggested that, while the most recent image is typically the most influential, prior imaging still provides additional prognostic value.

cs.CV↗

arXiv2024

Fréchet Video Motion Distance: A Metric for Evaluating Motion Consistency in Videos

Significant advancements have been made in video generative models recently. Unlike image generation, video generation presents greater challenges, requiring not only generating high-quality frames but also ensuring temporal consistency across these frames. Despite the impressive progress, research on metrics for evaluating the quality of generated videos, especially concerning temporal and motion consistency, remains underexplored. To bridge this research gap, we propose Fréchet Video Motion Distance (FVMD) metric, which focuses on evaluating motion consistency in video generation. Specifically, we design explicit motion features based on key point tracking, and then measure the similarity between these features via the Fréchet distance. We conduct sensitivity analysis by injecting noise into real videos to verify the effectiveness of FVMD. Further, we carry out a large-scale human study, demonstrating that our metric effectively detects temporal noise and aligns better with human perceptions of generated video quality than existing metrics. Additionally, our motion features can consistently improve the performance of Video Quality Assessment (VQA) models, indicating that our approach is also applicable to unary video quality evaluation. Code is available at https://github.com/ljh0v0/FMD-frechet-motion-distance.

cs.CV↗

arXiv2024

SwinGNN: Rethinking Permutation Invariance in Diffusion Models for Graph Generation

Diffusion models based on permutation-equivariant networks can learn permutation-invariant distributions for graph data. However, in comparison to their non-invariant counterparts, we have found that these invariant models encounter greater learning challenges since 1) their effective target distributions exhibit more modes; 2) their optimal one-step denoising scores are the score functions of Gaussian mixtures with more components. Motivated by this analysis, we propose a non-invariant diffusion model, called $\textit{SwinGNN}$, which employs an efficient edge-to-edge 2-WL message passing network and utilizes shifted window based self-attention inspired by SwinTransformers. Further, through systematic ablations, we identify several critical training and sampling techniques that significantly improve the sample quality of graph generation. At last, we introduce a simple post-processing trick, $\textit{i.e.}$, randomly permuting the generated graphs, which provably converts any graph generative model to a permutation-invariant one. Extensive experiments on synthetic and real-world protein and molecule datasets show that our SwinGNN achieves state-of-the-art performances. Our code is released at https://github.com/qiyan98/SwinGNN.

cs.LG↗

arXiv2024

The number of quasi-trees of bouquets with exactly one non-orientable loop

Recently, Merino extended the classical relation between the $2n$-th Fibonacci number and the number of spanning trees of the $n$-fan graph to ribbon graphs, and established a relation between the $n$-associated Mersenne number and the number of quasi-trees of the $n$-wheel ribbon graph. Moreover, Merino posed a problem of finding the Lucas numbers as the number of spanning quasi-trees of a family of ribbon graphs. In this paper, we solve the problem and give the Matrix-Quasi-tree Theorem for a bouquet with exactly one non-orientable loop. Furthermore, this theorem is used to verify that the number of quasi-trees of some classes of bouquets is closely related to the Fibonacci and Lucas numbers. We also give alternative proofs of the number of quasi-trees of these bouquets by using the deletion-contraction relations of ribbon graphs.

math.CO↗

arXiv2024

AutoCast++: Enhancing World Event Prediction with Zero-shot Ranking-based Context Retrieval

Machine-based prediction of real-world events is garnering attention due to its potential for informed decision-making. Whereas traditional forecasting predominantly hinges on structured data like time-series, recent breakthroughs in language models enable predictions using unstructured text. In particular, (Zou et al., 2022) unveils AutoCast, a new benchmark that employs news articles for answering forecasting queries. Nevertheless, existing methods still trail behind human performance. The cornerstone of accurate forecasting, we argue, lies in identifying a concise, yet rich subset of news snippets from a vast corpus. With this motivation, we introduce AutoCast++, a zero-shot ranking-based context retrieval system, tailored to sift through expansive news document collections for event forecasting. Our approach first re-ranks articles based on zero-shot question-passage relevance, honing in on semantically pertinent news. Following this, the chosen articles are subjected to zero-shot summarization to attain succinct context. Leveraging a pre-trained language model, we conduct both the relevance evaluation and article summarization without needing domain-specific training. Notably, recent articles can sometimes be at odds with preceding ones due to new facts or unanticipated incidents, leading to fluctuating temporal dynamics. To tackle this, our re-ranking mechanism gives preference to more recent articles, and we further regularize the multi-passage representation learning to align with human forecaster responses made on different dates. Empirical results underscore marked improvements across multiple metrics, improving the performance for multiple-choice questions (MCQ) by 48% and true/false (TF) questions by up to 8%. Code is available at https://github.com/BorealisAI/Autocast-plus-plus.

cs.LG↗

arXiv2024

Partial-dual polynomial as a framed weight system

Recently, Chmutov proved that the partial-dual polynomial considered as a function on chord diagrams satisfies the four-term relations. In this paper, we show that this function on framed chord diagrams also satisfies the four-term relations, i.e., is a framed weight system.

math.CO↗

arXiv2024

Twist polynomial as a weight system for set systems

Recently, Chmutov proved that the partial-dual polynomial considered as a function on chord diagrams satisfies the four-term relation. Deng et al. then proved that this function on framed chord diagrams also satisfies the four-term relation, i.e., is a framed weight system. In this paper, we extend their results to the twist polynomial of a set system by proving that the twist polynomial on set systems satisfies the four-term relation.

math.CO↗

arXiv2024

Eulerian-minors and a concise recursive characterization of 4-regular planar graphs

An Eulerian-minor of an Eulerian graph is obtained from an Eulerian subgraph of the Eulerian graph by contraction. The Eulerian-minor operation preserves Eulerian properties of graphs and moreover Eulerian graphs are well-quasi-ordered under Eulerian-minor relation. In this paper, we characterize Eulerian, planar and outer-planar Eulerian graphs by means of excluded Eulerian-minors, and provide a concise recursive characterization to 4-regular planar graphs.

math.CO↗

arXiv2023

What Constitutes Good Contrastive Learning in Time-Series Forecasting?

In recent years, the introduction of self-supervised contrastive learning (SSCL) has demonstrated remarkable improvements in representation learning across various domains, including natural language processing and computer vision. By leveraging the inherent benefits of self-supervision, SSCL enables the pre-training of representation models using vast amounts of unlabeled data. Despite these advances, there remains a significant gap in understanding the impact of different SSCL strategies on time series forecasting performance, as well as the specific benefits that SSCL can bring. This paper aims to address these gaps by conducting a comprehensive analysis of the effectiveness of various training variables, including different SSCL algorithms, learning strategies, model architectures, and their interplay. Additionally, to gain deeper insights into the improvements brought about by SSCL in the context of time-series forecasting, a qualitative analysis of the empirical receptive field is performed. Through our experiments, we demonstrate that the end-to-end training of a Transformer model using the Mean Squared Error (MSE) loss and SSCL emerges as the most effective approach in time series forecasting. Notably, the incorporation of the contrastive objective enables the model to prioritize more pertinent information for forecasting, such as scale and periodic relationships. These findings contribute to a better understanding of the benefits of SSCL in time series forecasting and provide valuable insights for future research in this area. Our codes are available at https://github.com/chiyuzhang94/contrastive_learning_time-series_e2e.

cs.LG↗

arXiv2023

Rapid FRD determination for multiplexed fibre systems -- I. The quasi-near field model and its uncertainties

Focal Ratio Degradation (FRD) in fibres is a crucial factor to control in astronomical instruments in order to minimize light loss. As astronomical instrumentation has advanced, the integration of large populations of fibres has become common. However, determining FRD in multiplexed fibre systems has become a challenging and time-consuming task. The Integral Field Unit for the Fiber Arrayed Solar Optical Telescope (FASOT-IFU) represents the most densely arranged fibre-based IFU in a single unit. Due to the close packing of fibres in the V-groove of the slit end, measuring FRD is particularly challenging as the output spots are prone to overlapping with adjacent fibres. In this paper, a novel method based on the quasi-near field model is proposed to enable rapid FRD measurement in highly multiplexed fibre systems like IFUs and multi-object observation systems. The principle and uncertainties associated with the method are investigated. The method's validity is demonstrated by applying it to determine the FRD in FASOT-IFU, with the achieved FRD performance meeting the acceptable requirements of FASOT-IFU, where the output focal ratio primarily falls within the range of 5.0-7.0. The results indicate that the proposed method offers several advantages, including the simultaneous and rapid measurement of FRD in multiple fibres with high accuracy (error smaller than 0.35 in F-ratio). Furthermore, besides FRD, the method exhibits potential for extensive measurements of throughput, scrambling, and spectral analysis.

astro-ph.IM↗

arXiv2023

Alignment of the Alpha Magnetic Spectrometer (AMS) in space

The Alpha Magnetic Spectrometer (AMS) is a precision particle physics detector operating at an altitude of 410 km aboard the International Space Station. The AMS silicon tracker, together with the permanent magnet, measures the rigidity (momentum/charge) of cosmic rays in the range from 0.5 GV to several TV. In order to have accurate rigidity measurements, the positions of more than 2000 tracker modules have to be determined at the micron level by an alignment procedure. The tracker was first aligned using the 400 GeV/c proton test beam at CERN and then re-aligned using cosmic-ray events after being launched into space. A unique method to align the permanent magnetic spectrometer for a space experiment is presented. The developed underlying mathematical algorithm is discussed in detail.

hep-ex↗

arXiv2022

Partial-twuality polynomials of delta-matroids

Gross, Mansour and Tucker introduced the partial-twuality polynomial of a ribbon graph. Chumutov and Vignes-Tourneret posed a problem: it would be interesting to know whether the partial duality polynomial and the related conjectures would make sense for general delta-matroids. In this paper we consider analogues of partial-twuality polynomials for delta-matroids. Various possible properties of partial-twuality polynomials of set systems are studied. We discuss the numerical implications of partial-twualities on a single element and prove that the intersection graphs can determine the partial-twuality polynomials of bouquets and normal binary delta-matroids, respectively. Finally, we give a characterization of vf-safe delta-matroids whose partial-twuality polynomials have only one term.

math.CO↗

arXiv2022

Hierarchically Constrained Adaptive Ad Exposure in Feeds

A contemporary feed application usually provides blended results of organic items and sponsored items~(ads) to users. Conventionally, ads are exposed at fixed positions. Such a static exposure strategy is inefficient due to ignoring users' personalized preferences towards ads. To this end, adaptive ad exposure has become an appealing strategy to boost the overall performance of the feed. However, existing approaches to implementing the adaptive ad exposure still suffer from several limitations: 1) they usually fall into sub-optimal solutions because of only focusing on request-level optimization without consideration of the long-term application-level performance and constraints, 2) they neglect the necessity of keeping the game-theoretical properties of ad auctions, which may lead to anarchy in bidding, and 3) they can hardly be deployed in large-scale applications due to high computational complexity. In this paper, we focus on long-term performance optimization under hierarchical constraints in feeds and formulate the adaptive ad exposure as a Dynamic Knapsack Problem. We propose an effective approach: Hierarchically Constrained Adaptive Ad Exposure~(HCA2E). We present that HCA2E possesses desired game-theoretical properties, computational efficiency, and performance robustness. Comprehensive offline and online experiments on a leading e-commerce application demonstrate the significant performance superiority of HCA2E over representative baselines. HCA2E has also been deployed on this application to serve millions of daily users.

cs.AI↗

arXiv2022

Twist monomials of binary delta-matroids

Recently, we introduced the twist polynomials of delta-matroids and gave a characterization of even normal binary delta-matroids whose twist polynomials have only one term and posed a problem: what would happen for odd binary delta-matroids? In this paper, we show that a normal binary delta-matroid whose twist polynomials have only one term if and only if each connected component of the intersection graph of the delta-matroid is either a complete graph of odd order or a single vertex with a loop.

math.CO↗

arXiv2022

CrossLoc: Scalable Aerial Localization Assisted by Multimodal Synthetic Data

We present a visual localization system that learns to estimate camera poses in the real world with the help of synthetic data. Despite significant progress in recent years, most learning-based approaches to visual localization target at a single domain and require a dense database of geo-tagged images to function well. To mitigate the data scarcity issue and improve the scalability of the neural localization models, we introduce TOPO-DataGen, a versatile synthetic data generation tool that traverses smoothly between the real and virtual world, hinged on the geographic camera viewpoint. New large-scale sim-to-real benchmark datasets are proposed to showcase and evaluate the utility of the said synthetic data. Our experiments reveal that synthetic data generically enhances the neural network performance on real data. Furthermore, we introduce CrossLoc, a cross-modal visual representation learning approach to pose estimation that makes full use of the scene coordinate ground truth via self-supervision. Without any extra data, CrossLoc significantly outperforms the state-of-the-art methods and achieves substantially higher real-data sample efficiency. Our code and datasets are all available at https://crossloc.github.io/.

cs.CV↗

arXiv2021

Counterexamples to the interpolating conjecture on partial-dual genus polynomials of ribbon graphs

Gross, Mansour and Tucker introduced the partial-dual orientable genus polynomial and the partial-dual Euler genus polynomial. They showed that the partial-dual genus polynomial for an orientable ribbon graph is interpolating and gave an analogous conjecture: The partial-dual Euler-genus polynomial for any non-orientable ribbon graph is interpolating. In this paper, we first give some counterexamples to the conjecture. Then motivated by these counterexamples, we further find two infinite classes of counterexamples.

math.CO↗

arXiv2021

Social NCE: Contrastive Learning of Socially-aware Motion Representations

Learning socially-aware motion representations is at the core of recent advances in multi-agent problems, such as human motion forecasting and robot navigation in crowds. Despite promising progress, existing representations learned with neural networks still struggle to generalize in closed-loop predictions (e.g., output colliding trajectories). This issue largely arises from the non-i.i.d. nature of sequential prediction in conjunction with ill-distributed training data. Intuitively, if the training data only comes from human behaviors in safe spaces, i.e., from "positive" examples, it is difficult for learning algorithms to capture the notion of "negative" examples like collisions. In this work, we aim to address this issue by explicitly modeling negative examples through self-supervision: (i) we introduce a social contrastive loss that regularizes the extracted motion representation by discerning the ground-truth positive events from synthetic negative ones; (ii) we construct informative negative samples based on our prior knowledge of rare but dangerous circumstances. Our method substantially reduces the collision rates of recent trajectory forecasting, behavioral cloning and reinforcement learning algorithms, outperforming state-of-the-art methods on several benchmarks. Our code is available at https://github.com/vita-epfl/social-nce.

cs.LG↗

arXiv2021

Twist polynomials of delta-matroids

Recently, Gross, Mansour and Tucker introduced the partial duality polynomial of a ribbon graph and posed a conjecture that there is no orientable ribbon graph whose partial duality polynomial has only one non-constant term. We found an infinite family of counterexamples for the conjecture and showed that essentially these are the only counterexamples. This is also obtained independently by Chumutov and Vignes-Tourneret and they posed a problem: it would be interesting to know whether the partial duality polynomial and the related conjectures would make sence for general delta-matroids. In this paper, we show that partial duality polynomials have delta-matroid analogues. We introduce the twist polynomials of delta-matroids and discuss its basic properties for delta-matroids. We give a characterization of even normal binary delta-matroids whose twist polynomials have only one term and then prove that the twist polynomial of a normal binary delta-matroid contains non-zero constant term if and only if its intersection graph is bipartite.

math.CO↗