Unlearning Bias
ID: HR0011ST2025D-03Type: BOTH
Overview

Topic

Unlearning Bias

Agency

Agency: DODBranch: DARPA

Program

Type: STTRPhase: BOTH

Topic Link

https://www.dodsbirsttr.mil/topics-app/

Additional Information

https://www.dodsbirsttr.mil/topics-app/
Timeline
    Description

    The Department of Defense, through DARPA, is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Unlearning Bias" in artificial intelligence (AI) models. The objective is to develop innovative methods to detect and remove bias from generative models without needing access to the original training data, with an emphasis on defining targeted biases, creating de-biasing methods, and establishing evaluation criteria to ensure effectiveness. This initiative is crucial for advancing responsible AI practices, as bias in AI systems poses significant challenges for both government and industry applications. Proposals are due by February 26, 2025, with the opportunity opening on January 29, 2025, and further details can be found at the official solicitation link: https://www.dodsbirsttr.mil/topics-app/.

    Files
    Title
    Posted
    PDF.html
    It seems that there is no text provided to summarize. Please provide the text you would like me to summarize.
    Similar Opportunities
    Mitigating Explicit and Implicit Bias Through Hybrid AI -
    DOD
    The Department of Defense, through DARPA, is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Mitigating Explicit and Implicit Bias Through Hybrid AI." The objective is to develop innovative neuro-symbolic methods that address both explicit biases, defined through natural language, and implicit biases that arise from data and model parameters in artificial intelligence systems. This initiative is critical as it aims to enhance the safety and reliability of generative AI technologies, which are increasingly recognized for their potential to perpetuate biases, thereby impacting both government and civilian applications. Proposals are due by February 26, 2025, with the opportunity opening on January 29, 2025, and further details can be found at the official solicitation link: https://www.dodsbirsttr.mil/topics-app/.
    Model-Agnostic Detection of Bias (MAD-Bias) -
    DOD
    The Department of Defense, through DARPA, is seeking innovative proposals for the Small Business Innovation Research (SBIR) program focused on the Model-Agnostic Detection of Bias (MAD-Bias). The objective is to develop new methodologies for detecting bias in datasets independent of the AI models that utilize them, employing techniques such as geometric and topological data analysis. This initiative is crucial for addressing bias in AI systems, which poses significant challenges for responsible AI adoption across both government and industry sectors. Proposals for Phase I are due by February 26, 2025, with the opportunity opening on January 29, 2025, and further details available at the official solicitation link: https://www.dodsbirsttr.mil/topics-app/.
    Bias Datasets for the Department of Defense (DoD) -
    DOD
    The Department of Defense (DoD), through the Defense Advanced Research Projects Agency (DARPA), is seeking proposals for the development of bias datasets specifically tailored for evaluating artificial intelligence (AI) models. The primary objective is to create new datasets that can effectively detect bias in DoD applications, alongside a baseline model for comparison and proposed metrics for evaluation. This initiative is crucial as it addresses the pervasive issue of bias in AI, which hinders responsible AI adoption and operational effectiveness within the DoD and industry. Interested parties should note that the solicitation is set to open on January 29, 2025, with proposals due by February 26, 2025. For more information, please visit the official source link at DoD SBIR/STTR.
    Analyzing the Trade-off Between Bias and Accuracy -
    DOD
    The Department of Defense, through DARPA, is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Analyzing the Trade-off Between Bias and Accuracy" in artificial intelligence (AI) systems. The objective is to develop novel methods for measuring and validating the trade-offs between accuracy and fairness in AI models, particularly in DoD-relevant use cases. This initiative is critical as it addresses the pervasive issue of bias in AI, which impacts both government and industry applications, including the de-biasing of generative models. Proposals for Phase I are due by February 26, 2025, with the opportunity opening on January 29, 2025, and further details can be found at the official solicitation link: https://www.dodsbirsttr.mil/topics-app/.
    Addressing Underrepresentation Through Data Augmentation -
    DOD
    The Department of Defense, through the Defense Advanced Research Projects Agency (DARPA), is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Addressing Underrepresentation Through Data Augmentation." The objective is to develop innovative data augmentation techniques that mitigate biases in artificial intelligence (AI) models, particularly those stemming from socio-ethno-political disparities. This initiative is crucial for enhancing the robustness and fairness of AI applications, which are increasingly vital across various government and civilian sectors. Proposals for Phase I are due by February 26, 2025, with a focus on demonstrating proof-of-concept technologies, while Phase II will involve the development and demonstration of a prototype system. Interested parties can find more information and submit proposals through the official solicitation link: DoD SBIR/STTR.
    Artificial Intelligence for Interoperability -
    DOD
    The Department of Defense (DoD) is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Artificial Intelligence for Interoperability," specifically aimed at enhancing data interoperability within tactical environments. The objective is to apply Large Language Models (LLMs) and other AI approaches to automate and support the integration of disparate warfighter systems, addressing challenges related to data unification and interoperability across various formats and systems. This initiative is crucial for improving operational efficiency and reducing redundancy in military systems, ultimately supporting the warfighter's mission. Phase I proposals are invited with a funding limit of $250,000 for a six-month performance period, with the application due by February 26, 2025. For more information, interested parties can visit the solicitation link at DoD SBIR.
    Artificial Intelligence for Interoperability -
    DOD
    The Department of Defense (DoD) is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Artificial Intelligence for Interoperability," aimed at enhancing data integration across various systems within tactical environments. The primary objective is to apply Large Language Models (LLMs) and other AI methodologies to address challenges related to data unification and interoperability, ensuring reliable performance regardless of system or data format. This initiative is crucial for improving operational efficiency and reducing redundancy in warfighting systems, ultimately supporting the Army's national AI leadership goals. Interested parties can submit Phase I proposals with a funding limit of $250,000 for a six-month performance period, with the application due by February 26, 2025. For more details, visit the solicitation agency's website at DoD SBIR/STTR.
    Radar Signal Processing Improvements for Probability of Detection -
    DOD
    The Department of Defense, specifically the Army, is seeking innovative solutions through the Small Business Innovation Research (SBIR) program to enhance radar signal processing for improved probability of detection against evolving long-range threats. The primary objective is to develop novel signal processing techniques that can be integrated into existing ground-based radar systems without compromising current scan times. This initiative is critical for maintaining national security and enhancing the capabilities of radar systems across various applications, including drone detection, weather monitoring, and healthcare. Interested parties can submit Phase I proposals with a funding limit of $250,000 for a six-month feasibility study, with the application due by February 26, 2025. For more information, visit the official solicitation page at DOD SBIR.
    Radar Signal Processing Improvements for Probability of Detection -
    DOD
    The Department of Defense, specifically the Army, is seeking innovative solutions to enhance radar signal processing capabilities to improve the probability of detection for longer-range threats without compromising current scan times. The objective is to develop novel signal processing techniques that can be integrated into existing ground-based radar systems, with an emphasis on conducting a feasibility study during Phase I, which includes laboratory simulations and a test plan. This initiative is critical for advancing radar technology, with potential applications in various fields such as drone detection, weather monitoring, and healthcare. Proposals for Phase I are due by February 26, 2025, with a funding limit of $250,000 for a six-month performance period. Interested parties can find more information and submit proposals through the official solicitation link at https://www.dodsbirsttr.mil/topics-app/.
    Prophylactics/Therapeutics for Neurotoxin Exposure -
    DOD
    The Department of Defense (DOD) is seeking innovative solutions through the Small Business Innovation Research (SBIR) program to develop prophylactics and therapeutics for neurotoxin exposure, specifically targeting organophosphorus compounds used in chemical warfare. The objective is to create a biological agent that can prevent illness from exposure to G-type and V-type nerve agents, with a focus on stability, efficacy, and pharmacokinetics for both prophylactic and therapeutic applications. This initiative is crucial for enhancing the safety of military personnel on the battlefield and protecting civilians from potential chemical attacks. Interested parties should note that the solicitation will open on January 29, 2025, and close on February 26, 2025, with further details available at the DOD SBIR website.