Linyun He work email
- Valid
- Valid
Linyun He personal email
Linyun He phone numbers
Currently work as a machine learning engineer in Meta.Was a staff machine learning scientist in VISA research.Ph. D from Washington University in St. Louis.Have extensive working (as a data scientist) and research experiences in large scale statistical modeling, data analysis, optimization problems and machine learning methods.
-
Machine Learning EngineerPlaidAustin, Tx, Us -
Machine Learning EngineerMeta Dec 2022 - PresentMenlo Park, Ca, Us -
Staff Machine Learning ScientistVisa Jun 2019 - Dec 2022Foster City, California, UsMachine/Deep Learning Anomaly Detection Graph DetectionPaaS Time-series Forecasting Fraud/Risk Detection -
Graduate Research AssistantWashington University In St. Louis Aug 2017 - May 2019St. Louis, Mo, Us★𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐀𝐩𝐩𝐫𝐨𝐱𝐢𝐦𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐓𝐡𝐞𝐫𝐦𝐚𝐥 𝐈𝐧𝐯𝐞𝐫𝐬𝐞 𝐅𝐮𝐧𝐜𝐭𝐢𝐨𝐧 (> 2.6 um).- Applied the neural network to approximate the inverse function of a complicated thermal model (no closed-form expression).- Utilized the linear generator, simple autoencoder, variational autoencoder, and 𝐆𝐀𝐍s (Generative Adversarial Networks) to generate more training samples based on limited measurements.- Built a theoretical framework in general for all thermal spectrometers (easy, medium, hard and crazy).- Implemented in MATLAB and Python (𝐓𝐞𝐧𝐬𝐨𝐫𝐅𝐥𝐨𝐰 and Scikit-learn).- Achieved 900% more accuracy for THEMIS (hard spectrometer) from 9% error to 1%.- Achieved breaking-trough for CRISM (crazy spectrometer) from ”impossible” to 1.8% error. -
Teaching AssistanceWashington University In St. Louis Sep 2017 - Dec 2017St. Louis, Mo, UsWorked as a teaching assistance for the course "Digital Signal Processing".
-
Graduate Research AssistantWashington University In St. Louis Aug 2015 - Jul 2017St. Louis, Mo, Us★𝐇𝐲𝐩𝐞𝐫𝐬𝐩𝐞𝐜𝐭𝐫𝐚𝐥 𝐈𝐦𝐚𝐠𝐞 𝐑𝐞𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐃𝐞𝐧𝐨𝐢𝐬𝐢𝐧𝐠. - Developed a 𝐬𝐭𝐚𝐭𝐢𝐬𝐭𝐢𝐜𝐚𝐥 𝐡𝐲𝐩𝐨𝐭𝐡𝐞𝐬𝐢𝐬 𝐭𝐞𝐬𝐭 based on limited quantity of data.- Modeled the problem as a large-scaled penalized optimization from the 𝐁𝐚𝐲𝐞𝐬𝐢𝐚𝐧 𝐦𝐞𝐭𝐡𝐨𝐝.- Achieved 55% better spatial resolution and 90% more accuracy than traditional Gaussian denoising methods.- Performance: denoising spectra much better than other non-statistical methods and statistical methods with improper assumptions. Simulations show that relative errors are decreased to 0.002% because our hypothesis method can always select the correct model.
-
Data Science - Geospatial Co-OpMonsanto Company Jan 2018 - Jun 2018- Developed a model to predict corn seeds supply in micro-sizes.- Selected features from different datasets based on the significance.- Implemented in Python, R and collaborated in GIThub with supply chain, Global IT and business groups.- Reduced prediction errors from 70% down to 10.2%, directly leading to $2,000,000 budget reduction per year.- Poster presentation on 2018 Technology Community of Monsanto.
Linyun He Skills
Linyun He Education Details
-
Washington University In St. LouisElectrical And Electronics Engineering -
Washington University In St. LouisEngineering Data Analytics And Statistics
-
University Of Electronic Science And Technology Of ChinaElectrical And Electronics Engineering
Frequently Asked Questions about Linyun He
What company does Linyun He work for?
Linyun He works for Plaid
What is Linyun He's role at the current company?
Linyun He's current role is Machine Learning Engineer.
What is Linyun He's email address?
Linyun He's email address is jh****@****isa.com
What is Linyun He's direct phone number?
Linyun He's direct phone number is +131430*****
What schools did Linyun He attend?
Linyun He attended Washington University In St. Louis, Washington University In St. Louis, University Of Electronic Science And Technology Of China.
What skills is Linyun He known for?
Linyun He has skills like Matlab, C, Latex, Microsoft Office, Algorithms, C++, Programming, Simulations, Java, Data Analysis, Mysql, Python.
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial