GLSearch: Unsupervised Maximum Common Subgraph Detection

We propose GLSEARCH, a Graph Neural Net- work (GNN) based learning to search model. Our model is built upon the branch and bound algo- rithm, which selects one pair of nodes from the two input graphs to expand at a time. Instead of using heuristics, we propose a novel GNN-based Deep Q-Network (DQN), which leverages search information and graph coarsening, to select the node pair, making the search process faster and more adaptive. We also leverage the search process to provide supervision in a pre-training and imitation learning.

Bai, Y.*, Xu, D.*, Sun, Y., & Wang, W. (2021, July). GLSearch: Maximum Common Subgraph Detection via Learning to Search. In International Conference on Machine Learning (pp. 588-598). PMLR. [paper link]

NeuralMCS: Supervised Maximum Common Subgraph Detection

We propose an efficient MCS detection algorithm, NeuralMCS, consisting of a novel neural network model that learns the node-node correspondence from the ground-truth MCS result, and a subgraph extraction procedure that uses the neural network output as guidance for final MCS prediction, with a hierarchical loss function capturing equivalent ground truths at different granularities.

Bai, Y.*, Xu, D.*, Gu, K., Wu, X., Marinovic, A., Ro, C., Sun, Y., & Wang, W. (2019). Neural maximum common subgraph detection with guided subgraph extraction. [paper link]

Embedded Machine Learning on STM Sensortile

I designed and implemented a real-time motion recognition system using CNNs and LSTMs to classify 9DOF signals, with heavy data augmentation through additive Gaussian noise and signal shifting. The system used BLE communication to relay sensor data to compute resources, and digital signal processing via band-pass filtering and boxcar smoothing to clean raw data.

[Tutorials link][GitHub link]

Uber Technologies (Summer 2021)

Machine Learning Engineering Intern

I worked on an information theoretic approach to AutoML, and implemented a functioning highly-scalable model and feature selection system that identified optimal geospatial features for production.

Uber Technologies (Summer 2020)

Software Engineering Intern

I analyzed pricing policies with the data science team and implemented a new end-to-end product in Python, Java, Go, JS, and SQL.

NASA Jet Propulsion Laboratory (Summer 2019)

Software Engineering Intern

I designed and developed a new programming language for compiling and verifying sattelite command files then ingesting them into a database.

NASA Jet Propulsion Laboratory (Summer 2018)

Software Engineering Intern

I developed a reverse data compilation engine to verify data packets between mission control and satellites.

CS30: Introduction to Computer Science (Summer 2021)

I set up the final project, a coding competition, covering various search algorithms: BFS, DFS, Best-First Search, Djikstra's, and A*.

E96C: Introduction to Engineering Design (2017-2020)

I helped debug software and hardware errors for students developing an embedded machine learning motion recognition system.