Haw-minn Lu

Haw-minn Lu, Adrian Kwong, Jose Unpingco

Jupyter has become the go-to platform for developing data applications but data and security concerns, especially when dealing with healthcare, have become paramount for many institutions and applications dealing with sensitive information. How then can we continue to enjoy the data analysis and machine learning opportunities provided by Jupyter and the Python ecosystem while guaranteeing auditable compliance with security and privacy concerns? We will describe the architecture and implementation of a cloud based platform based on Jupyter that integrates with Amazon Web Services (AWS) and uses containerized services without exposing the platform to the vulnerabilities present in Kubernetes and JupyterHub. This architecture addresses the HIPAA requirements to ensure both security and privacy of data. The architecture uses an AWS service to provide JSON Web Tokens (JWT) for authentication as well as network control. Furthermore, our architecture enables secure collaboration and sharing of Jupyter notebooks. Even though our platform is focused on Jupyter notebooks and JupyterLab, it also supports R-Studio and bespoke applications that share the same authentication mechanisms. Further, the platform can be extended to other cloud services other than AWS.

Haw-minn Lu

Most machine learning models, especially artificial neural networks, require numerical, not categorical data. We briefly describe the advantages and disadvantages of common encoding schemes. For example, one-hot encoding is commonly used for attributes with a few unrelated categories and word embeddings for attributes with many related categories (e.g., words). Neither is suitable for encoding attributes with many unrelated categories, such as diagnosis codes in healthcare applications. Application of one-hot encoding for diagnosis codes, for example, can result in extremely high dimensionality with low sample size problems or artificially induce machine learning artifacts, not to mention the explosion of computing resources needed. Quasi-orthonormal encoding (QOE) fills the gap. We briefly show how QOE compares to one-hot encoding. We provide example code of how to implement QOE using popular ML libraries such as Tensorflow and PyTorch and a demonstration of QOE to MNIST handwriting samples.

Haw-minn Lu, Giancarlo Perrone, José Unpingco

Although data may be abundant, complete data is less so, due to missing columns or rows. This missingness undermines the performance of downstream data products that either omit incomplete cases or create derived completed data for subsequent processing. Appropriately managing missing data is required in order to fully exploit and correctly use data. We propose a Multiple Imputation model using Denoising Autoencoders to learn the internal representation of data. Furthermore, we use the novel mechanisms of Metamorphic Truth and Imputation Feedback to maintain statistical integrity of attributes and eliminate bias in the learning process. Our approach explores the effects of imputation on various missingness mechanisms and patterns of missing data, outperforming other methods in many standard test cases.