DARPARA2401.pdf

DARPARA2401 YOUNG FACULTY AWARD (YFA) 1

Research Announcement Young Faculty Award (YFA)

Defense Sciences Office

DARPARA2401

November 7, 2023

DARPARA2401 YOUNG FACULTY AWARD (YFA) 2

Table of Contents I. Funding Opportunity Description

A. Introduction B. Program Description/Scope C. Program Structure D. Topic Areas (TAs)

II. Award Information A. General Award Information B. Fundamental Research

III. Eligibility Information A. Eligible Applicants B. Organizational Conflicts of Interest C. Cost Sharing/Matching

IV. Application and Submission Information A. Address to Request Application Package B. Content and Form of Application Submission C. Submission Dates and Times D. Funding Restrictions E. Other Submission Requirements

V. Application Review Information A. Evaluation Criteria B. Review and Selection Process C. Countering Foreign Influence Program (CFIP) D. Federal Awardee Performance and Integrity Information (FAPIIS)

VI. Award Administration Information A. Selection Notices B. Administrative and National Policy Requirements C. Reporting

VII. Agency Contacts VIII. Other Information

A. Program Overview Slides B. Frequently Asked Questions (FAQs) C. Collaborative Efforts/Teaming

ATTACHMENT A: EXECUTIVE SUMMARY TEMPLATE

ATTACHMENT B: PROPOSAL SUMMARY SLIDE TEMPLATE

ATTACHMENT C: PROPOSAL TEMPLATE VOLUME 1: TECHNICAL & MANAGEMENT

ATTACHMENT D: PROPOSAL TEMPLATE VOLUME 2: ADMINISTRATIVE & NATIONAL

POLICY REQUIREMENTS

DARPARA2401 YOUNG FACULTY AWARD (YFA) 3

PART I: OVERVIEW INFORMATION

Federal Agency Name: Defense Advanced Research Projects Agency (DARPA), Defense Sciences Office (DSO)

Funding Opportunity Title: Young Faculty Award (YFA)

Announcement Type: Initial Announcement

Funding Opportunity Number: DARPARA2401

Catalog of Federal Domestic Assistance (CFDA) Number(s): 12.910 Research and Technology Development

Dates (All times listed herein are Eastern Time.)

o Posting Date: November 7, 2023 o Executive Summary Due Date: December 13, 2023, 4:00 p.m.

o FAQ Submission Deadline: February 15, 2024, 4:00 p.m. See Section VIII.B.

o Full Proposal Due Date: February 22, 2024, 4:00 p.m.

Anticipated Individual Awards: DARPA anticipates multiple awards.

Anticipated Funding Available for Award: Each award will include a 24-month base period (a maximum of $500,000) and a 12-month option period (a maximum of $500,000).

Types of Instruments that May be Awarded: Grants

Agency Contacts

Technical POC: Rohith Chandrasekar, Program Manager, DARPA/DSO

RA Email: YFA2024@darpa.mil

RA Mailing Address:

DARPA/DSO

ATTN: DARPARA2401

675 North Randolph Street Arlington, VA 22203-2114

DARPA/DSO Opportunities Website: http://www.darpa.mil/work-with-us/opportunities

Teaming Information: See Section VIII.C for information on teaming opportunities.

Frequently Asked Questions (FAQ): FAQs for this solicitation may be viewed on the DARPA/DSO Opportunities Website. See Section VIII.B for further information.

mailto:YFA2024@darpa.mil http://www.darpa.mil/work-with-us/opportunities?tFilter=&oFilter=2&sort=name http://www.darpa.mil/work-with-us/opportunities?tFilter=&oFilter=2&sort=name

DARPARA2401 YOUNG FACULTY AWARD (YFA) 4

PART II: FULL TEXT OF ANNOUNCEMENT

I. Funding Opportunity Description

This Research Announcement (RA) constitutes a public notice of a competitive funding opportunity as described in 2 CFR § 200.203. Any resultant negotiations and/or awards will follow all laws and regulations applicable to the specific award instrument(s) available under this

RA.

A. Introduction The Defense Advanced Research Projects Agency (DARPA) Young Faculty Award (YFA) program aims to identify and engage rising stars in junior research positions in academia and equivalent positions at non-profit research institutions, particularly those without prior DARPA funding, to expose them to Department of Defense (DoD) needs and DARPA’s mission to create and prevent technological surprise. The YFA program will provide high-impact funding to elite researchers early in their careers to develop innovative new research that enables transformative DoD capabilities. Ultimately, the YFA program is developing the next generations of researchers focused on national security issues.

Before preparing an executive summary or proposal submission, proposers are encouraged to review the DARPA mission statement and current program descriptions at the DARPA website https://www.darpa.mil to familiarize themselves with examples of current DARPA investments.

This is not meant as instruction to duplicate those efforts, but rather to illustrate that current programs are aimed at research that will substantially advance our capabilities in these areas.

Once awards are made, each YFA program performer will be assigned a DARPA Program Manager with interests closely related to their research topic. The Program Manager will act as project manager and mentor to the YFA award recipients.

Proposers should also familiarize themselves with the “Heilmeier Catechism.” Details about the catechism and questions it seeks to answer can be found at https://www.darpa.mil/work-with-us/heilmeier-catechism.

B. Program Description/Scope DARPA is soliciting innovative research proposals in the areas of interest to DARPA’s six technical offices: Biological Technologies Office (BTO), Defense Sciences Office (DSO), Information Innovation Office (I2O), Microsystems Technology Office (MTO), Strategic Technology Office (STO), and Tactical Technology Office (TTO). Further detail regarding the specific technical areas of interest can be found under Section I.D “Topic Areas (TAs).”

Proposed research should investigate innovative approaches that enable revolutionary advances in science, devices, or systems. Specifically excluded is research that primarily results in evolutionary improvements to the existing state of practice.

Submissions responding to a TA under this RA should clearly describe the relevance and impact of the proposed concept(s) for addressing the national security challenges identified in the TA description (Section I.D); the current state-of-the-art technology; new insights to address the TA challenges; a credible research plan and schedule; and critical, quantitative milestones to be pursued over the research period.

This RA seeks grant proposals only. Submissions for any other award instrument type may be considered non-conforming with the RA and may not be reviewed.

https://www.darpa.mil/ https://www.darpa.mil/work-with-us/heilmeier-catechism https://www.darpa.mil/work-with-us/heilmeier-catechism

DARPARA2401 YOUNG FACULTY AWARD (YFA) 5

C. Program Structure This RA seeks grant proposals for a research activity consisting of a 24-month base period. No award type other than grants will be issued under this RA. Each 12-month interval of the base period shall not exceed $250,000. Proposals should also include a 12-month option period with a maximum funding level of $500,000. The 12-month option period, referred to as the “Director’s Fellowship,” will be reserved for a limited number of awardees who demonstrate exceptional YFA project performance over the 24-month base period. YFA recipients may request one No- Cost Extension (NCE) to the base period of performance (PoP) not to exceed 12 months. NCEs beyond 12 months for the base period require justification be submitted to the Agreements Officer for consideration. In order to be eligible for the Director’s Fellowship option period, the effort’s reporting must show they have expended a significant portion of received funding.

Director’s Fellowship nominations by the PM Mentor may be deferred until required reasonable expenditure levels have been reached, provided that deferment not surpass the Grant PoP, to include any requested extensions. Director’s Fellowship recipients may request one NCE for the Director’s Fellowship option PoP, not to exceed 12 months.

A target start date of August 2024 may be assumed for planning purposes.

As part of the program, a number of visits/exercises at a variety of DoD sites and facilities will be scheduled. These briefings and visits will provide YFA recipients unique, first-hand exposure to DoD personnel and technologies in the field; issues faced by the Military Services in execution of their missions; and current National Security challenges. It is expected that YFA recipients will participate in a subset of the visits/exercises made available to them. Participation in all such opportunities is not a requirement; however, lack of participation may impact the award of the Director’s Fellowship. Proposers are expected to include the necessary travel funds within the total budget of their proposal. For budgeting purposes, please plan for a minimum of 6 two-day meetings (three meetings in the Washington, D.C. area and three meetings in the San Francisco, CA area). Of the six meetings, four should occur over the course of the 24-month base period and two over the course of the 12-month option period. Travel proposed for non-DoD sponsored activities and events is highly discouraged, please see FAQ for more information.

D. Topic Areas (TAs) This RA solicits single principal investigator (PI) proposals for research and development in the specific TAs of interest articulated below. Prior to submitting a full proposal, proposers are strongly encouraged to first submit an executive summary as described in Section IV. At the executive summary phase, proposing PIs are limited to one executive summary per TA; at the full proposal phase, proposing PIs are limited to submitting only one full proposal to only one topic under this RA. Submitting more than one full proposal may result in all of the PI’s proposal submissions being determined non-conforming and being removed from award consideration.

Potential applicants are encouraged to carefully consider the descriptions of the TAs before submission. Each submission (executive summary or full proposal) must specify ONE and only one TA for the submission and identify this TA on the submission’s cover sheet. Executive Summaries and Full Proposals (limit of one proposal per proposer) that do not clearly address a specific topic may be deemed non-conforming and may not be reviewed. DARPA reserves the right to assign a proposal or an executive summary to a different topic area than indicated by the proposer.

DARPARA2401 YOUNG FACULTY AWARD (YFA) 6

Technical inquiries should be emailed to YFA2024@darpa.mil with the TA stated in the subject line. Your question will be distributed to the appropriate contact. Please see Section VIII.B for more details regarding the question and answer process.

1. 3D Microphysiologic Systems

DARPA is interested in advancing microphysiologic systems (MPS) to understand patient specific risks to environmental exposures or pharmaceutical interventions. Such systems, which can be composed of patient-specific cells, have the potential to model an individual’s physiologic risks to a large variety of threats. Many current MPS systems rely upon 2D models – layered sheets of cells. With advances in micropatterning and microfluidics of 2D substrates, MPS designers are able to tightly control the microenvironment of cultures – allowing more physiologic differentiation and “realistic” architectures. But, many tissues of interest have inherent 3D architectures – lungs, brain, liver, etc. Current 3D MPSs do allow induced pluripotent stem cells to self-develop into improved “realistic” microarchitectures – improving the ability to model tissues which are inherently 3D. However, in 3D MPSs, media, nutrients, growth factors and waste products passively diffuse – limiting the size of 3D cultures can grow as well as any spatial control of growth factor delivery necessary to promote a desired architecture. Thus, in order to improve 3D MPS to model in vivo tissues, two key challenges need to be addressed: 1) perfusion of nutrients and eliminating cellular waste evenly throughout the entire volume of MPS and 2) controlling the spatial perfusion of growth factors and physiologic active substances to 3D MPS. Strong proposals should address the challenges of fabricating biologically compatible systems which can distribute nutrients and waste removal throughout a 3D cell culture by year 1. By year 2, proposals should demonstrate spatial control of delivery of nutrient and incorporate living cell cultures into perfusion system.

2. Spatially Precise Microenvironments for Localized Therapeutic Intervention

Developing new therapeutic approaches to neuropathologies, such as traumatic brain injury (TBI), is of great interest to the Department of Defense (DoD) due to a heightened prevalence and severity in Service Members and Veterans. Pharmacologic therapies for TBI play important roles in acute treatment and post-acute care. However, many TBIs involve very specific brain regions and/or structures, making it difficult to target pharmacological/biologic therapeutic strategies to specific foci. For example, small molecules are designed to interact with specific brain receptors, yet these receptor targets are often more broadly expressed and not limited solely to the damaged region. This “off-target” effect often leads to significant and undesirable side effects. DARPA is seeking proposals on novel approaches for targeted, spatially precise therapeutic interventions in the brain. DARPA is not looking at the development of new drugs or implants per se, but to drive discovery of extremely localized therapeutics. Of particular interest is fundamental research that explores the underlying mechanisms of spatially constrained microenvironments by which alteration or tuning of brain cell/tissue function can be achieved.

Focusing on precision targeting after any type of brain injury or diagnosis of dysfunction, this effort differentiates itself from other current or previous DARPA programs involved in brain health and new therapeutic development within specific contexts. An increased understanding of these microenvironments will enable the development of advanced strategies and interventions for neurological disorders and dysfunctions (e.g., TBI, anxiety, depression) that can provide increased efficacy and mitigate off-target effects of traditional treatment.

mailto:YFA2024@darpa.mil

DARPARA2401 YOUNG FACULTY AWARD (YFA) 7

3. Rapid characterization and understanding of mechanisms of cell binding and entry by novel viral pathogens

The DoD has a need to develop effective medical countermeasures (MCMs) for the warfighter in a timely manner. The current COVID-19 pandemic has showcased the results of previous DoD investment in mRNA vaccine technology and rapid antibody discovery technologies from the ADEPT and Pandemic Prevention Platform (P3) programs, respectively. However, technological advances alone did not contribute to the swift development of vaccines and therapeutics against SARS-CoV-2 – the research and development community leveraged years of prior research establishing viral cell tropism and mechanism of viral entry by related coronaviruses (i.e. SARS- CoV-1, MERS-CoV). Previous research established that SARS-CoV-1 enters human epithelial cells via the ACE-2 receptor, and that the Spike protein generates the most robust immune response. Antibody discovery and manufacturing teams knew to screen for Spike-binding monoclonal antibodies, and test for the ability of their antibody candidates to block SARS-CoV- 2 binding to the ACE-2 receptor protein in vitro. If SARS-CoV-2 targeted different human cell types, or used a different receptor mechanism for cell entry, the MCM discovery and development process would have been significantly delayed by months to years. DARPA seeks to develop new, rapid techniques to screen viruses for both cell tropism and entry receptors. It is envisioned that the virus agnostic technologies will assess which human cells/tissues the virus enters and the receptor(s) to which it binds to gain entry. Additionally, approaches should achieve interpretable, actionable results in short timelines (days rather than months). Proposals should address:

1. Cell and/or tissue types (indicate source)

2. Assay(s) for assessing virus binding to cells and methods for determining the receptors

3. In vitro and/or in vivo validation of results

4. Virus agnostic nature of the methods – i.e. tested on different virus families

4. Outcome Selection Through Nontraditional Receptor Modulation

Understanding the mechanisms of action of toxicants is paramount to the chemical and biological defense programs of the DoD for both threat assessment and countermeasure development. Mixtures of fentanyl derivatives have been used as weapons, most notably in the 2002 Moscow Theater Siege, and drugs such as medetomidine have been researched by state actors for use as potential weapons. Current medical countermeasure (MCM) options for the treatment of opioid poisoning are limited to naloxone and its derivatives, which have several drawbacks. MCM options for the reversal of medetomidine effects are limited to veterinary use only and there are no options for humans. G-protein coupled receptors (GPCRs) are proteins that are the targets of many drugs including threat agents such as fentanyl and its derivatives, as well as other classes of toxins. GPCRs are membrane proteins that are essentially environmental sensors, communicating signals from the extracellular environment across the cell membrane to elicit a response inside the cell. A growing body of evidence suggests that GPCRs do not accomplish this cellular communication in a single way; receptor conformations or shapes can be modulated via ligand interactions at multiple sites that are unappreciated. Myriad possibilities for altering GPCR function exist that elicit both desired and adverse responses, be they based on interactions of the GPCR with auxiliary proteins, signaling pathway selectivity downstream of

DARPARA2401 YOUNG FACULTY AWARD (YFA) 8

the receptor, or binding at nontraditional sites by molecules of interest. DARPA is seeking proposals on work to develop molecules that act at GPCRs in nonclassical ways; namely those that leverage allostery, differential protein-protein interactions, and/or signaling pathway bias for utility against toxicant exposure relevant for national security. Proposals must provide rationale for how such development moves beyond state-of-the-art drugs that interact at classical sites to activate or inhibit GPCRs.

5. Biohybrid Additive Manufacturing of Strong, Flexible Optical Elements

This research topic aims to explore and build upon the advanced additive manufacturing landscape to develop the techniques to design, print, and test organic-hybrid tough and flexible optical elements, enabling future hardened and self-healing optical sensors for DoD applications.

Given the rapid development of additive manufacturing, it is now possible to create multiscale, multi-material, and multi-functional devices with varying dimensional scales spanning from the sub-micrometer to the millimeter. Recent research has demonstrated the printing of optically transparent fused silica nanostructures at low temperatures, which paves the way for incorporating biological and organic molecules into fabrication processes. Other recent work has shown that the multiscale lattice frameworks, inspired by the skeletal systems of deep-sea glass sponges, e.g., Euplectella aspergillum, can be printed, and the specific lattice design dictates the structure's mechanical properties. However, these efforts did not take advantage of the unique mechanical properties of E. aspergillum that result from the multiscale organization of inorganic silica with protein molecules. DARPA seeks to replicate the flexibility and self-healing properties of E. aspergillum fibers through improvements in additive manufacturing – integrating organic/biological and inorganic components, to improve fabrication speed and precision, and enhance multiscale 3D printing of flexible fiber optics by 1) successfully demonstrating the capability to use additive manufacturing to create bio-hybrid silica-based materials, and 2) demonstrating the ability to use additive manufacturing to mimic the multiscale architecture and mechanical and optical properties of E. aspergillum spicules.

6. Multiscale Modeling of Microbial Communities

This effort seeks to develop innovative computational approaches to model the complex dynamics of microbial communities. Specifically, this research aims to discover and apply novel multi-scale modeling approaches to inform and predict the dynamic behavior of bacteria in heterogeneous populations. Generalizable in silico models will help demystify complex and highly non-linear behavior of bacteria operating in microbial communities. Proposals should describe innovative approaches to model and predict quorum sensing in bacteria by considering heterogeneous phenotypes and variable environmental conditions. Computational models developed under this program should incorporate relevant biological processes including (but not limited to) the following: intra- and inter-cellular chemical communication; noise; gene expression and genomic (in)stability; gene regulatory networks (GSN); spatial-temporal heterogeneity; genetic and phenotypic heterogeneity. Performers under this topic will design and validate novel and generalizable computational multiscale models that will fundamentally improve our understanding regarding the complex dynamic behavior of microbial communities.

7. Decentralized Foundation Models from Small Data

Self-supervised pre-trained foundation models have recently found tremendous successes in a wide range of applications in natural language and computer vision problems (e.g., BERT, GPT, MoCo, and CLIP). These pre-trained foundation models rely on a variety of masking patterns to

DARPARA2401 YOUNG FACULTY AWARD (YFA) 9

construct “next-word-prediction-type” self-supervised training, thus enabling the training on massive-scale unlabeled datasets. Interestingly, the “next-word-prediction-type” self-supervised training bears striking similarity with sequential decision-making tasks typically seen in reinforcement learning (RL). Motivated by this connection, there have been some recent attempts on using self-supervised pre-trained foundation models for RL. However, the use of self-supervised pretrained foundation models has remained limited to conventional centralized single-agent RL. On the other hand, in complex tactical environments, many of the sequential decision-making problems are by nature multi-agent reinforcement learning (MARL) problems and fundamentally data-limited. So far, it remains unclear how to achieve the tremendous performance gains in MARL with self-supervised pre-trained foundation models, particularly in the data-limited regime. There are many significant and under-explored technical challenges in MARL when applying self-supervised pretrained transformer-based foundation models, including the designs of model architectures and self-supervised pre-training objectives, fundamental trade-offs between the choices of RL training paradigms, fully decentralized parameter-efficient finetuning and continual learning, etc. DARPA is seeking proposals on decentralized data-limited MARL with self-supervised pre-trained foundation models.

Successful proposals will enable applications of autonomy on large-scale networks (e.g., low-data swarm autonomy, ISR from multiple robots using vision-language models) and sample-efficient approaches for solving NP-hard combinatorial problems over networks. Proposals must include the development of theory and algorithms that address the fundamental challenges arising from limited data, computing, and networking constraints in MARL.

8. Compilation for Integrated Heterogeneous Systems

We are facing a “diversity crisis.” Moore’s law is no longer providing rapid improvements in performance, leading to an era in which specialization is king. Large DoD systems, such as F-35 avionic systems, increasingly comprise a heterogeneous ensemble of systems-on-chip (SoC) with specialized components, GPUs and accelerators, each with their own unique features, performance characteristics and instruction sets.

Ideally, these systems would be programmed in high level, domain-specific languages (DSLs) that abstract away these differences, leaving it to compilation tools to handle the details.

However, it is very time consuming and error prone to develop new compiler components that handle the diversity of hardware targets and source languages.

The goal of this research is to develop and integrate novel compiler components that address this diversity crisis by 1) Mapping elements of the source program to specific hardware targets; 2) Providing mechanisms, such as a variety of intermediate representations, to handle the diversity;

and 3) Performing optimizations at a variety of levels of abstraction. While recent research using machine learning (ML) techniques has begun to address compiler optimization challenges, proposers should address the technologies they will develop to address the above compiler challenges, what classes of diverse hardware they intend to address, and how they will verify that the compiled code is a faithful implementation of the source DSL.

9. FR (For Real) LLMs

Given the explosion in AI-generated content, especially through the use of large language models (LLMs), it is becoming increasingly critical to ground the models on scientific principles.

Though exceptionally powerful, current LLMs ‘learn’ from massive data, extracting purely statistical patterns without inherent consideration of known physical laws. This pure statistical

DARPARA2401 YOUNG FACULTY AWARD (YFA) 10

representation provides no guarantee that the model’s predictions are physically feasible or that they can appropriately generalize to unseen situations. The goal of this research program is to design new AI algorithmic approaches that incorporate scientific facts and physical principles into AI reasoning mechanisms. Examples of such principles can include laws of conservation, logical constraints, or statistical facts derived from relevant literature. This research should result in advancements and demonstrations of deep learning to ensure trustworthiness, especially in operational conditions that require generalization outside of previously observed conditions.

10. Causal and Mechanistic Explanations of Large Language Models

Much of the current state of the art in generative AI is based on large pre-trained models (LPTMs). Traditionally, neural networks of this type have been referred to as “black boxes,” and have been assumed to lack interpretable internal structure of the type that enables causal explanations and interventions regarding model behavior. Treating LLMs as inscrutable black boxes poses challenges for the safety and trustworthiness of such systems in that it fundamentally limits the types of approaches that can be pursued in order to audit or control systems in the wild, often favoring “guess and check” or “whack-a-mole” approaches to model development and refinement. Work in this topic area seeks to illuminate the basic representational and algorithmic building blocks in play within LPTMs, in order to facilitate finer-grained understanding and control of model behavior once deployed. Ideal outcomes will result in coherent human-understandable theories that explain how LPTMs work, why they behave the way they do, and how they will behave in new situations. Principled, theory-based approaches for controlling LPTMs in real world deployment will improve safety and trust in the black box nature of LPTMs.

11. Deployable Radiators for Small, High Power, Low Earth Orbit Spacecraft

Low earth orbit (LEO) spacecraft are of growing interest in various commercial and defense applications. These spacecrafts are exposed to multiple unique thermal control challenges, including a rapid rotation period (as low as ~90 minutes), increased aerodynamic drag due to proximity to earth, and material degradation due to interaction with atomic oxygen. They are subject to widely-varying internal and external thermal loads during orbit, requiring careful thermal control to insure proper operation of the deployed microsystems. Enhancing the capabilities of spacecraft-deployed microsystems results in larger heat dissipations, which must be effectively transmitted out of the spacecraft and rejected by the radiators. Small spacecraft such as CubeSat, a cube of 10 cm side (or 1U) or its multiples, are increasingly attractive, but are limited in spacecraft surface area for heat rejection. Deployable radiators allow for significant increase in heat removal capabilities, provided low thermal resistance links from the spacecraft also can be enabled. This topic seeks proposals on innovative deployable radiator concepts that allow >200 W of heat rejection capability for 6U LEO spacecraft. The targeted operating temperature range is -40 oC to 75 oC, mass per unit surface area < 6 kg/m2, thickness < 0.6 cm, specific stiffness >100 and specific thermal conductivity >5. Proposed approaches must be supported via underlying modeling and experimental validation.

12. Optical Coupler Arrays

Foundry processes for photonic integrated circuits (PICs) have enabled larger and more complex photonic circuits for a variety of applications including sensing, signal processing and interconnects. This scaling also has increased the quantity of off-chip optical couplers. Current approaches to coupling take up valuable chip real estate, have limited optical bandwidth, and are

DARPARA2401 YOUNG FACULTY AWARD (YFA) 11

high-loss which negatively impacts the PIC performance. The development of coupling technology similar to the ball grid array technology for electronics would allow for the simplification of PIC chips and enable the 3D integration of PICs with other microelectronic chips. This topic is soliciting proposals for the development of small-area, low-loss (<0.5 dB) optical coupling technology for high-density coupling arrays between foundry fabricated PICs.

Proposed technology should be compatible with current PIC foundry platforms, enable scaling to large number of connections, and have >1 µm misalignment tolerance.

13. Photonic Integrated Generators of Multiphoton Entanglement

The ability to create complex entangled states of light is central to the quantum technologies promising to transform computing, communication, and sensing. While laboratory experiments have demonstrated exotic forms of quantum light such as NOON states, cat states, and Greenberger–Horne–Zeilinger (GHZ) states, obtaining quantum technology advantages requires overcoming key challenges in scalability, quality, and generation rate. The rise of nonlinear integrated photonics in recent years presents opportunities to not only miniaturize the discrete-optical hardware required to generate entanglement, but also exceed multiple performance attributes owing to enhanced nonlinear efficiency in chip-scale optical devices and the capacity to multiplex large numbers of modes in complex photonic circuits. This topic seeks to develop foundational capabilities in multiphoton entangled state generation using integrated photonics.

Proposals should identify the innovations in photonic devices and circuits necessary to prepare and verify multiphoton entanglement, and include a constructive plan to experimentally demonstrate quantum light states at baseline rate > 1 MHz and fidelity > 75%. Concepts that generate large photon number states (>> 2), significantly exceed baseline rate and fidelity goals, and/or support emerging architectures in quantum information and measurement are of particular interest.

14. Hardware/Software Fault Injection

Fault injection, also known as glitching, is a cyber threat that controls hardware by introducing faults into the system. For example, researchers in 2015 were able to bypass game console security protections by resetting the chip at a specific time, thereby preventing specific instructions from executing. Today, finding and patching fault injection vulnerabilities require exquisite knowledge of the system under test which does not scale to the diversity of modern systems. While researchers have proposed software-only mitigations for a subset of glitching attacks, DARPA is interested in understanding the full breadth of fault injection attack threats and potential strategies for mitigating against them in future systems beyond general purpose processors and the associated software. This topic seeks proposals for research that can distill the problem space into fundamental properties, such as timing as mentioned above, as well fundamental requirements, such as the specific instructions mentioned above, so as to understand the research space. Techniques that focus only on elucidating the fundamental properties or only on elucidating the fundamental requirements are out of scope. Finding, preventing, and patching fault injection vulnerabilities requires knowledge of both the fundamental properties and requirements. Proposals that only focus on finding new vulnerabilities without the fundamental theories or requirements are also out of scope. Proposals must include an approach for ensuring the verifiability of discovered theories as well as repeatability and reproducibility of research results.

15. Advanced Structural Energetics

DARPARA2401 YOUNG FACULTY AWARD (YFA) 12

Developing advanced structural energetics for rocket propellants is of great national interest for both military and rapidly growing commercial launch services. Composite energetics have shown little improvement in mechanical properties or processability in the preceding decades, limiting grain geometries and rate of production. Conversely, the analogous structural polymer matrix composites have found critical uses in everything from consumer goods to military hardware. This ubiquity is built on decades of materials research to enable higher strength and durability while simultaneously decreasing cost and increase rate of production. DARPA is seeking proposals to bridge the material science of structural polymer chemistry and processing with energetics to enable future advances in energetics formulation and fabrication. New approaches to analysis and testing of cross-linking, interfacial bonding, and load transfer are encouraged. Proposals must include both modeling of energetic performance and experimental validation of mechanical properties. Where appropriate, surrogate non-energetic materials may be used for experimental testing when properly validated.

16. Beyond line-of-sight Power Beaming

Laser-enabled Optical power beaming is a potentially transformative technology for lunar or space-based power distribution and communication. However, this technology is fundamentally limited by line of sight. In the lunar environment, the horizon is <2 km away, limiting the area of regard of an optical power distribution node and requiring widely-distributed, mass-intensive relay systems to share power or communications “over the horizon”.

Recent work may change this emerging paradigm. Atomic and molecular polarization give rise to optical forces that trap particles (optical tweezers) and affect the refractive index of light propagation. Engineered mutual interactions between an expanding particle beam and a diffracting light beam can produce a self-channeling effect,1 enhancing the propagation distance beyond that determined by the Rayleigh criterion for a diffracting beam in vacuum. It is possible that this enables the optimization of self-guided particle and light beams such that they remain collimated over the horizon, enabling a revolutionary change to the art of the possible with power beaming to remote receivers beyond line of sight.

DARPA seeks highly innovative proposals that will form the foundational knowledge to model scale-appropriate vacuum particle-light coupled beam propagation over multiple kilometers of distance. In particular, proposers must describe how their research could demonstrably contribute to enabling an over-the-horizon power beaming capability specific to the lunar environment, to include: schemes for optimal mutual guiding in the presence of locally varying gravitational accelerations on the Moon; other relevant loss, scattering or detrapping mechanisms; and preliminary estimates for efficiency and maximum propagation distances of proposed laser designs.

17. Gas Surface Interactions in Very Low Earth Orbit (VLEO)

Drag prediction and material survivability of spacecraft in VLEO are two key areas of research needed to support DoD operations in this relatively unexplored region of space. VLEO spacecraft may be designed with a variety of features including, but not limited to, solar panels, air-breathing inlets, antennas and multi-layer insulation. The drag and material survivability are areas linked by a common phenomenon: gas surface interactions. DARPA is seeking proposals focused on understanding the kinetic and chemical energy exchange between neutral diatomic

1 Castillo et al., 2022, Nature, DOI 10.1038/s41598-022-08802-z.

DARPARA2401 YOUNG FACULTY AWARD (YFA) 13

(N2 and O2) and atomic species (O) in addition to charged species (O2+, O+, NO+) given that a large portion of the ionosphere is coincident with VLEO. Research has investigated interactions with a fixed surface condition. For this YFA, energy exchange dependence on surface conditions, wall temperature and electric potential, are of particular interest with materials expected for VLEO spacecraft. Proposals ideally include a combination of modeling and experimental research results to address this topic. The resulting research is expected to enable improvements to orbital aerodynamic performance in VLEO.

18. Novel Underwater Propulsion

Traditionally, ships and submersibles use propellers, impellers, or waterjets as their main forms of propulsion to generate thrust. DARPA is interested in understanding the fundamental physics and engineering of novel propulsion to be able to develop future novel ship propulsion technologies. Specifically important are mechanisms to maximize efficiency of the propulsion system.

One possible avenue is bio-inspired propulsion. Fish and other undersea creatures use a combination of fins/appendages and kinematic motions to achieve high-speed and highly efficient locomotion as well as unique maneuvering capabilities. Examples of potential areas of research are advances in the fluid-structure interactions to minimize losses from the traveling wave along the structure, novel actuation mechanisms with large numbers of actuators to allow the structure to interact with the fluid to maximize efficiency, and novel control systems for large networks of sensors and actuators.

Possible avenues for investigation could include, but is not limited to, one of the following areas:

modeling and simulation (M&S) of the fluid dynamics and fluid-structure interactions of bio-inspired propulsion, development of biomechanics for propulsion, development of novel distributed sensing and actuation for soft bodies, M&S for the prediction of trades (performance, efficiency, size, weight, and power) compared to conventional propulsion, and development and extracting principles of sensorimotor control to simulate biological systems for propulsion.

19. Light-Matter hybrids in the Large Cooperativity Regime

The quantum coherent interplay between light and matter is a fundamental building block of a wide range of quantum phenomena of relevance to Quantum Information Science (QIS). At sufficiently strong interaction strengths, this interplay can give rise to entirely new phenomenology due to the hybridization between the optical and atomistic degrees of freedom.

Once the exclusive province of cold atom-based cavity quantum electrodynamic (QED) systems, this regime of strong light-matter interaction and large cooperativity is being attained in an increasingly disparate set of solid-state nanophotonic platforms due to the ability to confine optical excitations to spatial dimensions far below the wavelength of light and couple optical excitations to a wide range of atomic, molecular, electronic and phononic excitations. With suitably low optical losses and requisite levels of addressability, tunability, and quantum control, such solid-state light-matter hybrids can enable applications including new regimes of nanoscale quantum sensing, quantum transduction between disparate regions of the electromagnetic spectrum, quantum emulation of non-equilibrium quantum phenomena such as light-induced superconductivity, and nanoscale control of light-mediated chemical processes.

This research effort seeks to develop and validate solid-state and nanophotonic platforms capable of attaining the regime of large atom-light cooperativity, with potential applications to quantum

DARPARA2401 YOUNG FACULTY AWARD (YFA) 14

sensing, transduction and quantum emulation of non-equilibrium many-body quantum phenomena. Proposers should justify the proposed experimental platform and describe a research plan to develop and characterize their light-matter hybrid quantum system in order to demonstrate the regime of large cooperativity. The Year 3 option period may be used for the demonstration of QIS applications such as quantum-enhanced sensing, transduction or quantum emulation with the proposed hybrid quantum system.

20. Poetics of authenticity in human - AI generated content

This program seeks to develop quantitative metrics of authenticity and authentication in human and AI generated texts, expressions and interactions. A poetics of authenticity will explore content patterns that are grounded in human experiences at the individual, group and cross-cultural levels. AI systems can generate content and the authenticity and intents associated with this content can be opaque. Human expression changes over time based on stimuli and experiences, informing changes of expression of and receptivity to narratives and art. This now includes interaction with AI and human/AI generated expressions and art. This program seeks to develop quantitative metrics that support assessment and detection of authentic expression. This includes concepts of poetic resonances in groups and individuals, and metrics for assessment of the dynamics and co-evolution of poetic resonances and narratives in response to AI, AI/human and human generated content and stimuli. Proposers are expected to explore quantitative metrics for assessment and detection of authenticity in human, AI and human/AI generated content, including language based and bio-metric signatures. Successful proposals will articulate one or several approaches to quantitative authentication methodologies in Year 1 and will focus on the development of potential implementation paths in Year 2.

21. 3-Domain Integration

Modern fabrication technologies including 3-Dimension integration, bulk and surface micro machining and wafer bonding have enabled precise integration of 2 disparate domains on a single wafer platform. Examples of this include piezoelectric micro-electro-mechanical systems (piezoMEMS) and photonic integrated circuits, vapor cells and laser sources, etc. While engineering breakthroughs and yield in these highly desirable integration platforms remain outstanding challenges, this YFA focuses on “3-Domain” integration. DARPA is interested in revolutionary micro-fabrication technologies that bring three disparate domains {CMOS, Mechanical, photonics, spin wave, quantum, fluidic} on a wafer-manufacturable platform with the potential to disrupt the current integration bottlenecks. Successful proposals will identify approaches to achieving 3-Domain solutions with 3 or less critical, low-yield batch fabrication steps in Year 1, followed by successful fabrication of 3-Domain devices in Year 2. A potential Year 3 effort would be focused on demonstration of proposed 3-Domain devices towards applications that challenge traditional fabrication methods.

22. Electromagnetic Interference Safeguards Secured by Ionics (EMISSIons)

Electromagnetic interference, whether natural (e.g., solar storms, lightning) or manmade (e.g., power line surges, nuclear electromagnetic pulses) can damage or significantly degrade the performance of traditional electronics. Ion-based systems, however, are far more resilient to rapid changes in ambient electromagnetic fields. The primary reason for this resilience is that ionic motion is much more difficult to disrupt than electronic motion, since ions are several orders of magnitude more massive than electrons. The development of ionic and protonic devices—in analogy with electronic devices—has seen rapid expansion recently in a number of

DARPARA2401 YOUNG FACULTY AWARD (YFA) 15

applications, but the resilience of these systems to electromagnetic interference and radiation remains underexamined. DARPA is seeking proposals to develop new ionic and/or protonic circuits and devices. In particular, proposals should seek to understand the response of these devices to a range of electromagnetic interfering signals as well as their resilience in high-radiation environments. Proposals should include specific measures to evaluate the robustness of these circuits and their integration with traditional electronics in hybrid systems, with an eye toward potential deployment in mission critical systems.

23. Exploring the Limits of Reinforcement Learning Explainability

Reinforcement Learning has, in several ways, revolutionized the discovery of protocols for solving many scientific modeling and control tasks – in effect, discovering algorithms for performing these tasks. Yet the algorithms discovered, even when very effective, remain “hidden” in the platform (e.g. deep neural networks, kernels) that was employed in their RL discovery. We seek to unlock this description of what was found, and interpret/rationalize the successful algorithm discovered. We seek to understand whether we have rediscovered one of the ``traditional” theoretical/computational algorithms we always knew, something close to them, or something with fundamentally, qualitatively different capabilities. The path towards doing this envisaged by this call, is the attempt to construct – in a data driven way, and exploiting the suite of modern tools developed for ML tasks/architectures- transformations between the algorithm found and known algorithms. In the process of this construction, we want to discover whether (a) there exist “perfect”, global equivalencies between RL-discovered and known algorithms, suggesting we discovered something we already knew; or whether (b) the equivalence is only local/partial, and the transformation eventually ceases to exist/becomes singular. Studying the nature of this breakdown and the behavior in its neighborhood we can then acquire insight in the fundamental novelty -or in the fundamental shortcomings- of the RL-discovered algorithm. Ideally, it will help us rationally improve our traditional algorithmic toolkit.

This effort seeks to develop the beginnings of a framework of Interpretable Reinforcement Learning, leveraging current RL computational technologies, knowledge of established scientific computation algorithms, and data-driven tools for comparing them as outlined above. The investigator should demonstrate proofs of concept for algorithms arising in more than one domain (e.g. control, solution of differential equations, classification problems, data compression). Proposals should indicate a set of proposed test cases and a description of the data that will be collected/used for the effort, as well as benchmarks against which the work will be evaluated. Human subject research is outside the scope of this effort.

24. DSO Open YFA Topic

DSO invites proposers to submit innovative fundamental research concepts that address one or more of the following technical thrust areas: (1) Novel Materials & Structures, (2) Sensing & Measurement, (3) Computation & Processing, (4) Enabling Operations, (5) Collective Intelligence, and (6) Emerging Threats. A description of each of these thrust areas is described below and includes a list of example research topics that highlight several (but not all) potential areas of interest. Proposals must investigate innovative approaches that enable revolutionary advances. DSO is explicitly interested in nascent high-risk, high-reward research relevant to the Department of Defense.

Novel Materials & Structures: This thrust includes, but is not limited to, science and

DARPARA2401 YOUNG FACULTY AWARD (YFA) 16

technology in quantum devices, atomic scale systems, and functional and structural materials.

Sensing & Measurement: This thrust includes, but is not limited to, science and technology in quantum sensing and metrology, seeing (sensing) the unseen, and novel light sources.

Computation & Processing: This thrust includes, but is not limited to, science and technology of quantum computing, cryptography, and modeling of complex systems.

Enabling Operations: This thrust includes, but is not limited to, technologies to support space-based operations, tactically remote environments, and resource assurance.

Collective Intelligence: This thrust includes, but is not limited to, exploration of complex social systems, adaptable Artificial Intelligence (AI), and AI-accelerated learning.

Emerging Threats: This thrust includes, but is not limited to, national security concerns related to global issues associated with raw material availability, environmental catastrophes, and digital societies.

II. Award Information A. General Award Information

DARPA anticipates multiple awards. The level of funding for individual awards made under this RA will depend on the quality of the proposals received and the availability of funds. Awards will be made to proposers2 whose proposals are determined to be the most advantageous to the Government, all evaluation factors considered. See Section V for further information.

The Government reserves…