Yu (Demi) Qin

New Orleans, LA 70118

Hello! Welcome to the space of 秦瑜, and you can call me Demi 👋

I’m a Ph.D. candidate in Computer Science at Tulane University, specializing in machine learning, topological data analysis, and scientific visualization. My dissertation, titled Metric Learning on Topological Descriptors is advised by Prof. Brian Summa and Prof. Carola Wenk.

In my research, I push the boundaries of how we understand and interpret complex datasets efficiently. I leverage machine learning and visualization to understand data structures and similarities from a topological perspective. I create models that not only measure data similarities at scale but also explain the ‘why’ behind these similarities. My work finds applications in diverse areas: enhancing medical imaging techniques for better diagnostic accuracy, advancing our understanding of climate models for more precise weather forecasting and climate change analysis, and increasing efficiency in supply chain management.

news

Apr 10, 2024	✨ New Paper! We introduce Merge Tree Neural Networks (MTNN), a leanred neural network model specifically designed for merge tree comparison. MTNN dramatically enhances the speed and quality of similarity computations,and speeds up the prior state-of-the-art by more than 100x on the benchmark datasets while maintaining an error rate below 0.1%, more details on [paper].
Dec 15, 2023	🛎 I’m attending NeurIPS 2023! I’m serving on the program committee for Symmetry and Geometry in Neural Representations (NeurReps) this year!
Oct 15, 2023	🎤 I’m attending IEEE VIS 2023! Check out our works in TopoInVis and EnergyVis. In TopoInVis, we propose a method to visualize topological importance [slide] , and in EnergyVis, we use Topological Data Analysis (TDA) to detect extreme climate events at scale [slide].
Sep 15, 2023	🏆 Selected as a Grace Hopper Celebration (GHC) 2023 scholar!

selected publications

A Domain-Oblivious Approach for Learning Concise Representations of Filtered Topological Spaces for Clustering

Yu Qin, Brittany Terese Fasy , Carola Wenk , and Brian Summa

IEEE Transactions on Visualization and Computer Graphics, 2021

Abs HTML

Persistence diagrams have been widely used to quantify the underlying features of filtered topological spaces in data visualization. In many applications, computing distances between diagrams is essential; however, computing these distances has been challenging due to the computational cost. In this paper, we propose a persistence diagram hashing framework that learns a binary code representation of persistence diagrams, which allows for fast computation of distances. This framework is built upon a generative adversarial network (GAN) with a diagram distance loss function to steer the learning process. Instead of using standard representations, we hash diagrams into binary codes, which have natural advantages in large-scale tasks. The training of this model is domain-oblivious in that it can be computed purely from synthetic, randomly created diagrams. As a consequence, our proposed method is directly applicable to various datasets without the need for retraining the model. These binary codes, when compared using fast Hamming distance, better maintain topological similarity properties between datasets than other vectorized representations. To evaluate this method, we apply our framework to the problem of diagram clustering and we compare the quality and performance of our approach to the state-of-the-art. In addition, we show the scalability of our approach on a dataset with 10k persistence diagrams, which is not possible with current techniques. Moreover, our experimental results demonstrate that our method is significantly faster with the potential of less memory usage, while retaining comparable or better quality comparisons.
Visualizing Topological Importance: A Class-Driven Approach

Yu Qin, Brittany Terese Fasy , Carola Wenk , and Brian Summa

In 2023 Topological Data Analysis and Visualization (TopoInVis) , 2023

HTML
Rapid and Precise Topological Comparison with Merge Tree Neural Networks

Yu Qin, Brittany Terese Fasy , Carola Wenk , and Brian Summa

2024

Abs HTML

Merge trees are a valuable tool in scientific visualization of scalar fields; however, current methods for merge tree comparisons are computationally expensive, primarily due to the exhaustive matching between tree nodes. To address this challenge, we introduce the merge tree neural networks (MTNN), a learned neural network model designed for merge tree comparison. The MTNN enables rapid and high-quality similarity computation. We first demonstrate how graph neural networks (GNNs), which emerged as an effective encoder for graphs, can be trained to produce embeddings of merge trees in vector spaces that enable efficient similarity comparison. Next, we formulate the novel MTNN model that further improves the similarity comparisons by integrating the tree and node embeddings with a new topological attention mechanism. We demonstrate the effectiveness of our model on real-world data in different domains and examine our model’s generalizability across various datasets. Our experimental analysis demonstrates our approach’s superiority in accuracy and efficiency. In particular, we speed up the prior state-of-the-art by more than 100x on the benchmark datasets while maintaining an error rate below 0.1%.