?

Artificial Intelligence Research & Development

Artificial Intelligence R&D

Keep up with the latest AI, ML and data science research from WWT

AI R&D at WWT is an applied research initiative focused on investigating the one- to three-year horizon of the artificial intelligence and machine learning (AI/ML) space.

Our team of data science and analytics experts develop, conduct and report on a wide range of internal projects grounded in our deep understanding of industry use cases.

Learn about our team and explore some of our work below.

Our People

AI R&D is an initiative formed by WWT individuals interested in the future of AI and ML. It functions as a rotational program with an Operations team and an R&D Working team composed of data scientists, data engineers and application engineers.


Our Platform

Our AI R&D platform is a cloud-based containerized environment that gives us the extensibility and elasticity to develop, build and test AI/ML solutions. The platform's flexible architecture optimizes network and storage, while GPU-enabled devices provide the technical capabilities for algorithm training. ML infrastructure tools are deployed on top of the AI R&D platform to streamline workflows and automate the productionalization of projects from dataset to deployment.

Our Projects

All AI R&D projects -- which are developed, funded and conducted internally -- are geared toward applied research that unearths scientific discoveries in the AI/ML space that are innovative, solve problems and have potential commercial application. Industry use cases and datasets -- from mining, motorsports, utilities and more -- power our AI R&D work.


Our Output

The goal of WWT's AI R&D program is to produce reusable platform, workflow and algorithm components that can be leveraged for future AI/ML project work or as demos in our Advanced Technology Center (ATC). All projects result in a white papers made accessible to our customers and the wider AI community. Explore our latest findings below.

AI R&D White Papers

Page Thumbnail
WWT Research

Smart Receiving: A Warehouse Automation Solution

WWT's Data Science team and Supply Chain Process Improvement team worked together on a Smart Receiving solution that leverages computer vision technologies in aiding operators in the material validation process at the receipt.
Page Thumbnail
WWT Research

Mitigating Bias in AI Using Debias-GAN

In this white paper, we propose a general framework, debias-GAN, to address possible bias in AI and Machine Learning (ML) algorithms by explicitly augmenting a training dataset for NLP models with underrepresented instances synthesized by a pretrained sequence generating model.
Page Thumbnail
ATC

Deploying the AI R&D MLOps Platform to Enable End-to-End ML Workflows

Deploying the Kubeflow MLOps platform in AWS to enabled our Data Science team to create end-to-end ML workflows for automated delivery of machine-learning models.
Page Thumbnail
ATC

MRI Radiomics: Tumor Identification and Reverse Image Search Using Convolutional Neural Network

This paper leverages deep learning to match MRI tumor images to scans with similar features with the goal of providing clinicians a tool to increase accuracy and speed to diagnosis.
Page Thumbnail
ATC

An Ensemble Approach to Data Mining for Real-time Information Retrieval

Learn about an approach to Information Retrieval (IR) using a combination of multiple Natural Language Processing (NLP) models.
Page Thumbnail
ATC

Machine Learning Models for Route Consolidation

We develop a generalizable machine learning method for route consolidation. The developed method is compared against a more traditional ad-hoc method. The machine learning method uses a deep autoencoder, K-means clustering and Procrustes distance. The machine learning method is shown to produce similar results to the more traditional method with the advantage of using a more generalizable approach.
Page Thumbnail
WWT Research

Obscuring and Analyzing Sensitive Information With Generative Adversarial Networks

In this white paper, we explore the feasibility of generating representative data for two types of data: binary input from medical records and real-valued sensor data from industrial mining trucks.
Page Thumbnail
ATC

Deep Learning With NVIDIA DGX-1

We highlight some practical considerations for the Deep Learning practitioner relevant to neural network training on the NVIDIA DGX-1.
Want more Data Analytics & AI content?