Karelia Pena

Lead Data Scientist · PhD, Electrical & Computer Engineering

I am a Lead Data Scientist at McGraw Hill, where I build AI solutions for education. Previously, I was a Senior Data Scientist on the Intuit Futures team, where I led Generative AI work on document intelligence and tax software development. That work led to two granted US patents, several pending applications, and the 2024 Intuit Platform Acceleration Group (PAG) Excellence Award.

I received my MSc and Ph.D. in Electrical and Computer Engineering from the University of Delaware (2020, 2023). Before that, I earned my B.S. in Electrical Engineering summa cum laude from the Universidad de Los Andes, Venezuela, the first woman to receive that distinction in the department’s 50+ year history.

My research interests include:

Graph Neural Networks
Graph signal processing
Natural language processing
Computer vision
Generative AI and large language models
Multimodal document intelligence and retrieval-augmented generation (RAG)
Hypergraph neural networks

NEWS:

Spring 2027: I will teach ELEG 604 Imaging and Deep Learning at the University of Delaware.
2026: Granted US Patent 12,639,341, "Identifying artifacts differences using graph neural networks."
2026: Patent applications published: "Context-aware information retrieval" and "Architecture and training method for multimodal content moderation model."
January 2026: Promoted to Lead Data Scientist at McGraw Hill.
2025: Patent application published: "Completeness graph generator."
October 2024: Joined McGraw Hill as a Senior Data Scientist.
2024: Our paper "T-HyperGNNs: Hypergraph Neural Networks via Tensor Representations" was published in IEEE Transactions on Neural Networks and Learning Systems.
2024: Two papers on compressive LiDAR were published in IEEE Transactions on Geoscience and Remote Sensing: "Transformer End-to-End Optimization of Compressive LiDARs Using Imaging Spectroscopy Side Information" and "HyperHeight LiDAR Compressive Sampling and Machine Learning Reconstruction of Forested Landscapes."
2024: Our paper on diabetic foot lesion segmentation with YOLOv5 and SAM was presented at IEEE ISBI 2024.
2024: Granted US Patent 11,861,884, "Systems and methods for training an information extraction transformer model architecture."
2024: Received the Intuit Platform Acceleration Group (PAG) Excellence Award.
December 2023: Our paper "Learning Hypergraphs Tensor Representations From Data via t-HGSP" was published in IEEE Transactions on Signal and Information Processing over Networks [DOI].
September 2023: Our paper "A Unified View Between Tensor Hypergraph Neural Networks and Signal Denoising" was presented at EUSIPCO 2023.
May 2023: Our paper "t-HGSP: Hypergraph Signal Processing Using t-Product Tensor Decompositions" was published in IEEE Transactions on Signal and Information Processing over Networks [DOI].
February 2023: Joined the Intuit Futures team as a Senior Data Scientist.
January 2023: Defended my Ph.D. dissertation, "Learning from High-Order Data: Hypergraphs and Smart Codes," at the University of Delaware.
September 2022: The preprint of our paper "t-HGSP: Hypergraph Signal Processing Using t-Product Tensor Decompositions" is available on TechRxiv [DOI].
June 2022: Our paper "HedcutDrawings: Rendering hedcut style portraits" has been accepted in the 33rd Eurographics Symposium on Rendering [DOI].
May 2022: Our research proposal on hypergraphs received the Defense Established Program to Stimulate Competitive Research (DEPSCoR) Award, receiving up to $600,000 over a three-year period.
May 2022: Our paper "Colored-QRNet: Fast QR Code Color Image Embedding" has been accepted in the 30th European Signal Processing Conference.
May 2022: I joined the Cortex team at Intuit as a Data Science Intern.
January 2022: Our paper "QRnet: Fast learning-based QR code image embedding" has been accepted for publication in Multimedia Tools and Applications [DOI].
December 2021: I got an offer to join Intuit as a Data Scientist Intern in the Summer of 2022.
August 2021: Our paper "Fast Spectral Clustering Method Based on Graph Similarity Matrix Completion" was accepted for publication in Elsevier Signal Processing [DOI].
June 2021: I was awarded the University of Delaware Doctoral Fellowship Award.
May 2021: I was awarded the Signals Processing and Communications Graduate Faculty Award.
October 2020: Our paper "Blue-Noise Sampling of Graph and Multigraph Signals: Dithering on Non-Euclidean Domains" was published in the IEEE Signal Processing Magazine [DOI].
September 2020: Invited talk on “Channel Coding Optimization for Visually Pleasant QR codes” at ITEC 2020, La Paz, Bolivia.
May 2020: Received a Master of Science degree from the University of Delaware.

Page updated

Google Sites

Report abuse