DHS awards $200K to begin automated machine learning prototype test for CBP’s GTAS
DataRobot received $200,000 to begin testing a prototype of a machine learning platform for Customs and Border Protection’s Global Travel Assessment System (GTAS) as part of S&T’s Silicon Valley Innovation Program (SVIP).
In March, S&T Homeland Security Advanced Research Project Agency Cybersecurity Division Director Douglas Maughan explained to the House Committee on Oversight and Government Reform’s Subcommittee on Information Technology, that, “CBP offers advanced passenger data-screening and targeting technology as an open source software project, known as the Global Travel Assessment System,” which “is a turn-key application that provides to CBP’s foreign counterpart agencies the necessary decision support system features to receive and store air traveler data, both Advanced Passenger Information (API) and Passenger Name Record (PNR), provide real-time risk assessment against this data based on a country’s own specific risk criteria and/or watch lists, and view high-risk travelers as well as their associated flight and reservation information.”
“The purpose of GTAS,” he said, “is to provide border security entities the basic capacity to ingest, process, query, and construct risk criteria against the industry-derived standardized air traveler information. The system provides border security organizations with the necessary tools to prescreen travelers entering into and leaving their countries.”
Maughan told the subcommittee, “DHS SVIP and CBP partnered to enhance the GTAS project with solutions from the global innovation community, namely new capabilities using AI and machine learning, and identifying the following three capabilities for consideration:
• Visualization: This would extend the basic flight and passenger tabular list screens with geospatial, link analysis, seat map visualization, or any other concepts that improve the software by presenting data graphically;
• Predictive Models: These would complement GTAS rules engine with statistical and machine learning models and a “predictive model engine” that performs real-time risk assessment; and
• Entity Resolution: This capability would enhance the basic name/date of birth and document matching algorithms to support more advanced entity identification and matching algorithms.
DHS said, “Developing predictive models in GTAS currently requires data scientists and specialists who work at a pace that risks a model being outdated by the time of its completion. DataRobot proposes to apply automated machine learning (AML) to GTAS to expedite the model development process,” adding, “Security standards can vary by country, potentially allowing unsafe items or travelers into American ports of entry. CBP provides the GTAS system to other countries free of charge to enable them to meet U.S. security standards through the analysis of risk criteria against standardized air traveler information.”
“With the number of international travelers to the United States increasing every year, we know we need better and faster tools to process incoming passengers,” said Melissa Ho, SVIP Managing Director. “An enhanced Global Travel Assessment System will mean a better travel experience for all passengers and increased safety for Americans.”
DHS said AML technology is far easier for non-data scientists to utilize than traditional machine learning because AML automatically performs “complicated modeling tasks and data preprocessing, which allows programs like GTAS to perform complex functions while simplifying user experience.”
“An AML platform could enable GTAS to produce increasingly accurate predictive models,” said Anil John, S&T Identity Management R&D Program Manager. “With a simplified user experience, non-data scientists, such as CBP officers, could have the ability to research, collaborate, test and deploy predictive algorithms and develop insights into potential threats.”
Under the Enhancements to the Global Travel Assessment System initiative:
• DataRobot Inc. was initially awarded a grant September 2017 to provide statistical and machine-learning models and a predictive/scoring engine that performs batch and real-time assessments in a fraction of the time it currently takes.
• Omniscience Corporation was selected to build a fully functional model manager that will allow DHS and partner organizations to build models that can take in a high number of dimensions and significantly large data sets. Its Initial award was in April 2017.
• Tamr was selected to use their core human-guided, machine learning software to achieve schema mapping and entity resolution. It received an initial award in December 2016.
As part of DHS’s Identity Management program, the Identity and Access Management Engine (IDAM-E) will also help the Homeland Security Enterprise (HSE) enable identity and access management solutions via stakeholder engagement, problem identification, projects, and research and development investments.
DHS says, “IDAM-E will bring expertise, technologies, tools, capabilities and approaches from government (U.S. & international) as well as external scientific, technical, industrial and academic sources to bear on identity, cyber and privacy problems identified by the Apex programs, IDAM-E, and DHS in general. When capabilities do not exist, build them via investments in research, prototypes, etc.”
IDAM-E key activities are:
• Leverage expertise and relationships that span the public and private sector to bring identity, information security and privacy capabilities to meet S&T program and HSE needs;
• Provide test-bed infrastructure and test and evaluation expertise to prototype, evaluate and validate technologies; and
• Make R&D investments to close technology gaps in areas of importance to the HSE.
According to S&T, “The Biometric Technology Engine (BT-E) establishes an enduring core capability by leveraging S&T’s biometric expertise and ensuring the re-use of biometric tools, methods, and best practices, as well as support robust testing and evaluation at the Maryland Test Facility (MdTF) to inform applications of biometric technology to specific operational use cases across Apex programs, S&T programs, DHS, and the HSE.”
It will also:
• Drive Efficiency: provide cross-cutting methods, best practices, and solutions to drive efficiencies across biometric programs;
• Address Needs: leverage combined capabilities of the Engine structure to address current, emerging, and future operational needs;
• Test and Evaluation: provide objective, first-class biometric testing and evaluation (T&E) services to Apex programs, S&T, DHS, and the HSE;
• Engage Industry: leverage combined industry insights and engage private sector to forage for innovative biometric solutions; and
• Encourage Innovation: drive biometric standards and innovation across the HSE.
The BT-E will also, DHS said, “accelerate effective integration of biometrics into operations and work in a cross-cutting fashion to mitigate operational inefficiencies.”
Companies participating in the SVIP are eligible for up to $800,000 in non-dilutive funding over four phases to adapt commercial technologies for homeland security use cases.
biometric exit | CBP | DHS | machine learning | passenger processing