There are no open Broad Agency Annoucnements at this time.

Program Manager

Torreon Creekmore

Program Information

IARPA-BAA-19-04

The goal of the SMART program is to automate the quantitative analysis of space-based imagery to perform broad-area search for natural and anthropogenic events and characterize their extent and progression in time and space. The SMART program aims to develop capabilities in the spectral and temporal domains, enabling seamless integration and fusion (i.e., absolute calibration) of data from multiple sensors to deliver a comprehensive representation of seven natural or anthropogenic evolving events. Examples of such events include: heavy construction, urban development, crop disease propagation, forest fire, insect or battle damage, human migration, mining, logging, farming, and other natural events such as flooding, mudslides, or earthquakes. The SMART program will require innovations in new computing approaches and calibration techniques in order to rapidly and reliably compare thousands of images from multiple sensors registered in space and time. The SMART program will also leverage algorithmic approaches to:

  1. Search for specific activities
  2. Detect and monitor activities throughout time and over broad areas
  3. Characterize the progression of events and activities temporally and categorically

Related Program(s)

CORE3D

Research Area(s)

  • Remote sensing
  • Image processing
  • Atmospheric correction
  • Machine learning
  • Radiometric calibration
  • Multi-spectral
  • Multi-temporal
  • Datacube
  • Data fusion
  • Automation
  • Spaceborne
  • Geo-registration
  • Change detection
  • Classification
  • Pixel-based

Related Article(s)

 

There are no open Research Opportunities at this time.

Program Manager

Jeffrey Alstott

Program Information

Broad Agency Announcement: W911NF-19-S-0012

TrojAI logo 72dpi 01Using current machine learning methods, an artificial intelligence (AI) is trained on data, learns relationships in that data, and then is deployed to the world to operate on new data. For example, an AI can be trained on images of traffic signs, learn what stop signs and speed limit signs look like, and then be deployed as part an autonomous car. The problem is that an adversary that can disrupt the training pipeline can insert Trojan behaviors into the AI. For example, an AI learning to distinguish traffic signs can be given just a few additional examples of stop signs with yellow squares on them, each labeled “speed limit sign.” If the AI were deployed in a self-driving car, an adversary could cause the car to run through the stop sign just by putting a sticky note on it. The goal of the TrojAI program is to combat such Trojan attacks by inspecting AIs for Trojans.

Research Area(s)

  • AI security
  • Trojan detection
  • Explainable AI

Related Article(s)

As part of its mission to address some of the most difficult challenges in the Intelligence Community, IARPA sponsors research programs and challenges that either leverage or improve Artificial Intelligence/Machine Learning (AI/ML), including:

IARPA is always seeking novel ideas aligned with our mission. If you are interested in working with IARPA through one of our existing solicitations, prize challenges, requests for information, or other mechanisms, please see this link for more details.