Illumination Works (ILW) is proud to announce the award of an Army Small Business Innovative Research (SBIR) Phase II contract to advance our Theia™ solution to a production-ready prototype. Theia provides artificial intelligence (AI) algorithms to perform automated data labeling and curation across many diverse data types and structures.
Theia is built using an ensemble machine learning (ML) approach, combining methods and models in the computer vision (CV), natural language processing (NLP), knowledge graph, and human-machine teaming domains. Phase II Theia development focuses on labeling and curating for the purpose of characterization of entities in the logistics community, but is modeled to be transferable across various organizations and industries.
Accelerating Information Retrieval for Knowledge & Intelligence
At present, Army initiatives gathering training data for AI require analysts to sift through volumes of data to manually tag and parse data for model training. Furthermore, building platforms that enable groups to locate relevant information and identify patterns requires subject matter experts (SME) to manually curate and characterize information, often times citing internal data, rather than utilizing the data within the platform itself. The Army seeks an AI-based automated and reliable data labeling and curation system that allows users to quickly search, filter, and select datasets for downstream analyses.
Through a Phase I SBIR, ILW demonstrated the ability of our Theia AI-driven solution to automatically label data through characterization using knowledge-based insights constructed from relevant data and authoritative sources. This Phase I SBIR proved this methodology speeds the time to identify data of relevance, improve downstream AI/ML with curated/pre-labeled data, and filter out data noise so analysts can focus on informative data to answer the questions at hand.
Advancing Theia to Production-Ready Prototype (TRL 6)
For the Phase II effort, ILW will enhance and optimize the build-out of a production-ready Theia prototype for insertion into an Army environment of choice in Phase III. This includes maturing Theia to accommodate automatic labeling and curation of the full swath of Army data sources and productionalize the capability for Army use, achieving a technical readiness level of 6 (TRL 6).
Key Differentiator of Theia
The key differentiator for Theia lies in its implementation of autonomous labeling, which programmatically traverses documents and data to precisely organize all entities and relationships to construct a knowledge retrieval system beyond a simple keyword search or phrase search engine. This system cycles back on itself to improve automated labeling capabilities and grow the universe of possible labeled entities, utilizing intelligent self-learning methods developed within Theia’s AI processing engine.
This Phase II award highlights the Army’s desire to be the tip of the spear in technology innovation.
Theia is built using truly groundbreaking and cutting-edge methods that will create a powerful solution for the Army. Theia will also forge ahead in the research and development paths within the AI community by implementing advancements on present day challenges, including self-learning modeling, zero- and one-shot learning, automated labeling, and knowledge graph data science. I personally am proud to be a part of the ILW team that gets to work on this innovation as I believe Theia will open the flood gates for knowledge-based AI solutions going forward.
John Tribble, Principal Data Scientist, Illumination Works
The AI/ML-driven automation powering Theia provides fast and accurate tagging and curation of multiple data sources, greatly reducing manual labeling and speeding information retrieval—providing knowledge and intelligence with applicability to many domains such as the intelligence, manufacturing, and government communities.
Improving the User Experience—Interactive Modeling & Analysis
The Theia technology developed during Phase II will provide users the ability to utilize labels for their own modeling and analysis. Labeling will be grown to include self-learning, semi-supervised methods and will be conducted over a plethora of sources. To assist users in leveraging these labels, a user interface will be developed into a usable prototype with a focus on retrieving documents and labels from the underlying label database.
As an enabling technology, Theia will greatly improve the efficiency of data organization, characterization, and retrieval to significantly reduce hands-on, manual tasks of Army data labeling teams. Theia’s AI-driven automation will empower the Army to speed information understanding—improving operational readiness and supporting the mission.
For more information, contact Jan Turkelson, Vice President, Government Division.
About Illumination Works
Illumination Works is a trusted technology partner in user-centric digital transformation, delivering impactful business results to clients through a wide range of services including big data information frameworks, data science, data visualization, and application/cloud development, all while focusing the approach on the end-user perspective.
Established in 2006, ILW has primary offices in Dayton and Cincinnati, Ohio, as well as physical operations in Utah and the National Capital Region. In 2020, Illumination Works adopted a hybrid work model and currently has employees in 14 states and is actively recruiting.
