HECTOR Proposers’ Day Brief Mark Heiligman Program Manager 26 July 2017 INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA HECTOR Program Proposers’ Day Agenda Time Topic 9 00 am – 9 30 am Registration and Check In 9 30 am – 9 45 am IARPA Overview and Remarks 9 45 am – 10 30 am HECTOR Program Overview 10 30 am – 11 00 am BAA Overview T E GFI GFE 11 00 am – 11 30 am Break 11 30 am – 12 00 pm Doing Business with IARPA 12 00 pm – 12 30 pm HECTOR Program Questions Answers 12 30 pm – 1 30 pm Lunch 1 30 pm – 3 00 pm Proposers' 5-minute Capability Presentations 3 00 pm – 4 00 pm Proposers’ Networking and Teaming Discussions Speaker IARPA management Mark Heiligman Program Manager Mark Heiligman Program Manager IARPA Acquisition Mark Heiligman Program Manager Attendees No Government Attendees No Government INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 2 Proposers’ Day Goals • Familiarize participants with IARPA and with the HECTOR program concept • Solicit feedback and questions • Foster networking and discussion of synergistic opportunities and capabilities among potential program participants A K A “teaming” • Please ask questions and make suggestions this is your chance to influence the design of the program • We appreciate and seek constructive feedback on any all aspects of the program design and program metrics • Record your questions and comments on the note cards provided and submit them to IARPA staff during the break • After today questions will be answered in writing on the program website • Once a BAA is released questions can only be submitted to the email address provided in the BAA INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 3 Disclaimer • These presentations are provided solely for information and planning purposes • The Proposers’ Day does not constitute a formal solicitation for proposals or abstracts • Nothing said at Proposers’ Day changes the requirements set forth in a BAA • A BAA supersedes anything presented or said by IARPA at the Proposers’ Day INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 4 This briefing is UNCLASSIFIED All slides are Prepub-Approved IARPA Overview Dr Bill Vanderlinde Chief Scientist July 2017 UNCLASSIFIED IARPA Mission and Method IARPA’s mission is to envision and lead high-risk high-payoff research that delivers innovative technology for future overwhelming intelligence advantage • Bring the best minds to bear on our problems – Full and open competition to the greatest possible extent funding scientists and engineers in academia and industry through contracts grants OTs and prize challenges – World-class rotational Program Managers • Define and execute research programs that – Have goals that are clear measureable ambitious and credible – Employ independent and rigorous Test Evaluation – Involve IC partners from start to finish – Run from three to five years – Publish peer-reviewed results and data to the greatest possible extent INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 6 UNCLASSIFIED IARPA’s Customers Central Intelligence Agency Department of State Defense Intelligence Agency National Security Agency Department of Energy National Geospatial-Intelligence Agency Department of the Treasury National Reconnaissance Office Drug Enforcement Administration Army Federal Bureau of Investigation Navy Department of Homeland Security Air Force Coast Guard Marine Corps INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 7 UNCLASSIFIED IARPA Highlights “One of the government’s most creative agencies ” – David Brooks NYT • Best known for quantum computing superconducting computing forecasting tournaments but our portfolio is diverse -- math CS physics chemistry biology neuroscience linguistics political science cognitive psychology “Everything from AI to Zika ” • Research highlights include – White House BRAIN Initiative National Strategic Computing Initiative – Nobel Prize for Physics – Science “Breakthrough of the Year” – MacArthur “Genius” – 2 000 journal articles • 70% of completed research transitioned to USG partners INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 8 UNCLASSIFIED Current IARPA Research Collection - Amon-Hen space SA - FELIX syn bio - FunGCAT syn bio - Ithildin chem detection - HFGeo HF geolocation - MAEGLIN CBRN - MOSAIC pattern of life - Odin biometrics - Proteos human ID - SILMARILS chem - SLiCE RF tracking - UnderWatch undersea - Seedlings and Studies Analysis Computing - Aladdin video search - Babel speech recognition - CORE3D 3D modeling - DIVA surveillance video - Finder geolocate images - Janus facial recog - KRNS neuroimaging - MATERIAL translation - SHARP training - Seedlings and Studies - C3 cryogenic computing - HECTOR encryption - LogiQ quantum - MICrONS neuromorphic - QEO quantum - RAVEN chip analysis - SuperTools cryogenic - TIC chip security - VirtUE cloud security - Seedlings and Studies Anticipatory Intel - CAUSE cyber I W - CREATE crowdsourcing - FUSE S T intel - Hybrid Forecasting I W - Mercury SIGINT I W - SCITE insider threats - Seedlings and Studies Prize Challenges - Nail-to-Nail Fingerprinting - Unconstrained Face Recognition - Functional Map of the World - MORGOTH’S CROWN INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 5 UNCLASSIFIED How to engage with IARPA • Website www IARPA gov – Reach out to us especially the IARPA PMs Contact information on the website – Schedule a visit if you are in the DC area or invite us to visit you • Opportunities to Engage – Research Programs • Multi-year research funding opportunities on specific topics • Proposers’ Days provide opportunities to learn what is coming and to influence programs – IARPA-Wide BAA “Seedlings” • Typically a 9-12 month study you can submit your research proposal at any time • Strongly encouraged informal discussion with a PM before proposal submission – Prize Challenges • No proposals required • Submit solutions to our problems if your solutions are the best you receive a cash prize and bragging rights – Requests for Information RFIs and Workshops • Provide input while IARPA is planning new programs INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 10 HECTOR Program Overview Mark Heiligman PM 26 July 2017 INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA Presentation Outline • • • • • • • • • • • • • Motivation and Objectives Current Status Program Approaches BAA Overview Program Structure and Deliverables Technical Milestones and Program Metrics GFI GFE and Test and Evaluation Reporting Requirements Schedule Management Plan and Teaming Eligibility Information Proposal Evaluation Criteria Program Summary INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 12 HECTOR Problem Compliance Solution Expanded Cryptographic Toolset • Protect Data Under Process • Homomorphic Encryption • Garbled Circuits Transparency Privacy • Functional Encryption • Zero-Knowledge Proofs • Verification of Computational Correctness Access • Secure Multi-party Computation Vision A System Development Platform • Allows decomposition of distributed systems • Allows specification analysis of security properties • Estimates resource costs of whole systems • New language allows expression of new concepts • Implementation of advanced cryptography • Verifies implementation of security properties • Automatically generates verification auditing tools Impact Transform Business Processes Systems to allow Mutually distrusting parties on untrusted computing platforms to collaborate on a shared computation for a result that all can trust in INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 13 What are you trying to do Transform access control and data protection through advanced cryptography and a holistic approach to systems engineering Today Limited cryptographic toolset and advanced cryptographic techniques are not expressible in today’s languages Systems developed using aggregation accretion integration followed by examining tweaking the security properties Security properties of systems are hard to tease out post-development Future HECTOR will streamline the development of large-scale distributed systems that make use of advanced cryptographic capabilities Architects security experts designers all have input to the process Impact Verifiable systems with trustable outputs in a malicious environment Exploring and mapping out the new security space Driving innovation while reducing costs INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 14 Why is it Hard • Overhead needs to be understood and reduced – In FHE 106 per multiplication is widely reported but orders of magnitude improvements have been seen in recent years • Each technology has its own improvement curve often independent of the others • But combining multiple new technologies concepts together imposes new costs • The overall cost to a large system is not obvious • Newer cryptographic frontiers are poorly understood • Particularly the use cases and security threat models • Academic assumptions may not match up to real-world threats • Retrofitting new concepts into old languages is a significant challenge • What are the security implications of design decisions • Need to build realistic systems to solve real-world problems at scale • Hard to explain the benefits and issues without a common language INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 15 How’s It Done Now Security of Data In Transit Symmetric cryptography was originally invented to protect sensitive data in transit across a hostile environment Once decrypted data protection would be lost INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 16 The Evolution of Cryptography Security of Data In Transit Asymmetric cryptography allowed communicants to “go secure” from an insecure state and to authenticate the source of a message Authentication of Data Source INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 17 The Evolution of Cryptography Security of Data In Transit Security of Data At Rest Cryptographic hashing and key derivation allowed for true protection of data at rest Authentication of Data Source Unfortunately once a device is unlocked just as for data in transit any data protection is lost INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 18 Next Steps in Cryptography Security of Data In Transit Security of Data At Rest Cryptographic computing techniques allow data to be processed without revealing its nature They allow fine-grained access control to the data itself and to the ability to perform computations on the data There is no loss of control Authentication of Data Source Security of Data In Process INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 19 Limitations of Present Approaches Today’s access control paradigm After each boundary failed we added another boundary without really fixing anything Facility Network Endpoint Adversaries Data • Currently no other way to protect data under process – Advanced crypto limited to academia and some pilots – Languages support at best one new concept at a time – Overhead costs decreasing but still potentially prohibitive INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 20 State-of-the-art Cryptography • Advanced technologies change the nature of computation – Fully Somewhat Partial Homomorphic Encryption – Verifiable Computing – Functional Encryption • Conditional Proxy Re-encryption – Zero-Knowledge Proofs – Oblivious RAM – Secure Multiparty Computation • Oblivious transfer Multiparty circuit evaluation • Private Set Intersection • Private Information Retrieval • Many of these can be brought into the mainstream See the Technology Primer slides for more details INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 21 UNCLASSIFIED Census Data Processing and Trade Space Evaluation EXAMPLE CHALLENGE #1 INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA UNCLASSIFIED 22 Census Data A • • • • B Case A researcher wants to perform computations on sensitive PII data held by the census bureau to get aggregate results Homomorphic Encryption Bob provides the data in an encrypted form Alice performs complex computations on the encrypted data and sends the encrypted result to Bob Bob decrypts the result and returns the unencrypted answer to Alice Verification Bob need to verify that the computation that Alice claims to have performed is actually the computation that Alice did perform Bob needs to check that the computation was in conformance with policy INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA One Problem Numerous Solutions Research Organization Census Bureau No Crypto Census Bureau Does All The Hard Work Program Policy Check Computation Data Results Functional Encryption A Priori Proof of Function Compliance Program Compliance Proof Results Computation Policy Check FunctionKey Data FHE Data Functional Encryption A Posteriori Proof of Computation Compliance Program Computation Compliance Proof FHE Results Results FHE Data Policy Check Data FunctionKey Decryption INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 24 Questions and Issues • Example Functional Encryption – The capability can be described very simply • I give you data but you can only evaluate functions that I authorize you to compute – However the underlying mechanism is inherently complicated – The security implications are also inherently complex – How accessible can we make the language syntax security reports • You can no longer simply “call” a function – there is an authorization process – Functional encryption is expensive • Are there theoretical lower limits to the overhead • When integrated into a large system is this still a bottleneck – Leakage is Poorly Understood • Similar questions and issues apply to other techniques INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 25 Algorithm Programming Language C 5 F-32 9 Programmed Algorithm Gate Compiler Key Security 128 Bits Crypto Ajtai-Dwork Security Settings Cryptography Choice Gate Compiled Algorithm Crypto-Compiler Homomorphically Encrypted Algorithm INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA Float C F J Input F J  F - 32 C  J 1 8 Output C Trusted Domain Untrusted Domain Encrypted Algorithm Algorithm Float C F J Input F J  F - 32 C  J 1 8 Output C HECTOR Compiler … A1  B1 B2 A2  B3 B4 A3  A1 A2 … Encryption Key Key QWERTYUIOP Private Data Data Encryption Encrypted Data Input ZXCVBNM BIG COMPUTER Temp 98 6 F Decryption Key Key QWERTYUIOP Private Data Encrypted Data Data Decryption Output ASDFGH Temp 37 C INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA System Development Platform System Specification System Design App Security Feasibility Study Implementation Framework Component Implementations Component ComponentImplementations Implementations Compiler Pre-built Libraries Ext Libraries API Translation Inputs Resource Estimation Report Intermediate Representation Linker Built Applications Verifier Verifiability Audit Tools Platform Outputs INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA What is new about HECTOR • Automation of flow-down design process – – – – Requirements constraints automatically verified Functional decomposition of distributed systems Performs resource estimation identifies bottlenecks Distills input from domain experts • System goals security requirements and threat environment are expressed once and drive system requirements and constraints • If any of these change the system design can be reassessed – Advanced crypto concepts are abstracted away • The capabilities they provide are built into an extensible language • The composition of multiple concepts is supported by language INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 29 What difference will HECTOR make • Reduce costs and risks associated with secure IC systems – An up-front assessments of the resource costs – Enable verifiable regulatory compliance greater partner collaboration • Deliver the infrastructure and tools to simplify security – Cryptographic implementations can be re-used in different contexts – Calculates security implications of architectural design decisions – Enables system certifiers to have an automated assurance of correctness possibly for the most critical aspect of a system • Empower system architects and software developers – Know what resources are required to get the job done – Intuitively develop secure large-scale distributed systems 7 26 2017 INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 1 Application 3 Database Query Telecom Database NSA FBI 4 Query Results FISA Court 2 Court Order John Q Public wants to know 1 Does the FISA application conform to policy and law Need to verify compliance with policy without compromising the content of the application a Encrypt FISA court application b Homomorphically compute whether the application is policy compliant 2 Is the FISA court applying the law correctly Need to verify that the FISA court order is limited to what was applied for a FISA court issues warrant in the form of a function key to allow a specific computation b Homomorphically compute whether the computation being warranted corresponds to the what was applied for 3 Is the database search being done correctly Need to verify that the function was applied correctly a Perform functional verification 4 Are the returns from the database search correct and are they legally compliant Need to confirm that the results conform to policy and law a Homomorphically compute whether the encrypted return is legally compliant INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA OFFICE DF THE NATIONAL INTELLIGENCE BAA Overview GFIIGFE BAA Highlights • Three core focus areas 1 System Development Platform 2 Programming Languages Representation Formats 3 Cryptographic Protocols and Optimization • Three phases Plan Design Implement Optimize Refine • Program Duration Five Years • Performers will demonstrate exemplar applications – – Initial challenge problems declared at start of program Some challenge problems defined by performers themselves INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA Core Focus Areas System Development Platform • Overview • The goal of the system development platform focus area is to implement a full suite of development tools for secure distributed applications • Most of the use cases targeted by the HECTOR program are inherently multiparty which could include distributed development and verification of securitycritical aspects of the system INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 34 Core Focus Areas Programming Languages Representation Formats • Overview • The goal of this focus area is to develop and document a number of programming languages and representation formats needed by the HECTOR program to develop exemplar applications that use these languages formats to meet challenge problems and to interface both with other focus areas within a performer team and with other performer teams with respect to standardization • One of the goals of HECTOR is the development of a component implementation language that allows for the intuitive expression of advanced cryptographic techniques and that allows for future expansion of its conceptual scope as new techniques are developed • Each offeror is expected to develop their own component implementation language which should be a full programming language that provides an extremely high level of abstraction around key management concepts and cryptographic operations INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 35 Core Focus Areas Cryptographic Protocols and Optimization Focus Area • Overview • The goal of this focus area is to implement or adapt existing cryptographic protocols so that they can be used within the system development platform in an interchangeable and composable fashion and to explore performance improvements INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 36 Collaboration and Standardization • While offerors are expected to develop their own unique tools and programming languages and research and develop their own optimizations to cryptographic protocols two key aspects of the HECTOR program would benefit from standardization and offerors will be expected to collaborate on a single common solution in those areas • One goal of HECTOR is to encourage standardization at the intermediate representation level so offerors will be expected to collaborate on a common representation language with coordination from the T E team • Offerors will also be expected to collaborate on a common standard module format for incorporating implementations of cryptographic protocols INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 37 Program Structure Phase I Phase I Planning and Design • Overview • The goal of Phase 1 is to allow performers to demonstrate the viability of their plans and develop a solid system design • During Phase I performers will develop detailed plans designs and specifications and build automated grammar and syntax checkers for any programming languages or representation formats developed during the year INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 38 Program Structure Phase I Phase I Plan and Design • Details • Implement GFI-specified protocols schemes selected to match challenge problems and perform initial baselining • Develop a detailed research plan for novel secure data services to include cryptographic computing concepts schemes and or protocols • Develop detailed specification documents and initial syntax checker utilities for performerspecific programming languages and representation formats • Collaboratively develop detailed specification documents for common standards to include the intermediate representation language and the metadata-rich module format • Develop syntax format checker utilities for the intermediate representation language and the metadata-rich module format • Develop a detailed software design document for the tools to be developed in Phase II that together form the system development platform • Develop exemplar application source artifacts to show how their programming languages and representation formats would be used to solve challenge problems associated with Phase I INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 39 Program Structure Phase II Phase II Implement • Overview • The goal of Phase II is to implement and demonstrate the full capabilities of the system development platform • Delivered software is expected to be “research-grade” rather than “productiongrade” but must be fully documented and capable of operating on and building applications for consumer-level computing hardware and operating systems INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 40 Program Structure Phase II Phase II Implement • Details • Research new concepts for secure data services • Research performance improvements in existing secure data services • Implement additional secure data services • Implement all system development platform functionality and tools • Develop exemplar applications to answer the challenge problems for Phase II INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 41 Program Structure Phase III Phase III Optimize and Refine • Overview • The goal of Phase III is to optimize and further increase the capabilities of the system development platform both in terms of its performance as a tool and in terms of the performance of the applications it can generate • To test the extensibility of the programming languages’ conceptual scope a new cryptographic computing concept will be introduced at the start of Phase III selected from those proposed by the performers and the test and evaluation team in Phase II • Each performer will be expected to demonstrate the inclusion of this new concept into their programming language and its mapping to functionality at the intermediate representation level INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 42 Program Structure Phase III Phase III Optimize and Refine • Details • Research and or implement performance improvements in existing secure data services • Implement additional secure data services • Incorporate optimization strategies into the compiler • Improve the accuracy and or performance of resource estimation tools • Develop exemplar applications to answer the challenge problems for Phase III INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 43 Out of Scope • Cryptanalysis of the cryptographic protocols schemes in particular cryptanalytic research into the hardness of properties labeled as hardness assumptions • Acquisition of high performance computers or equivalent hardware • Resource or security improvements that rely on trusted hardware • Development of special purpose hardware INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 44 HECTOR Deliverables • Languages Representation Formats – – – – – • Tools – – – – – • Concept-Extensible Application Implementation Language Annotated System Architecture Description Language Threat Model Security Model for Reasoning Engine Intermediate Representation Language Metadata-Rich Linkable Module Format System Design Tool Reasoning Engine for Security Feasibility Studies Compiler Linker Resource Estimator for Cryptographic Systems Automated Verifier Verification Tool Generator Advances in Cryptographic Frontiers – Implementations of New Existing Schemes and Protocols – New Concepts for Cryptographic Computing – Efficiency Improvements for Existing Concepts INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 45 Phase 1 HECTOR Deliverables Date Month 1 Month 3-4 Month 6 Month 10 Month 12 Month 12 Event Deliverable Program Phase I Program Kick-off Meeting Annual Site Visits  Year 1 Technical Exchange Meeting  TX1 First exchange on common standards Annual Principal Investigators  PI  Program Review Meeting  Year 1 Second exchange on common standards Performers deliver Month-12 Deliverables and Annual Research Report Concept-Extensible Application Implementation Language Annotated System Architecture Description Language Threat Model   Security Model Format for Reasoning Engine Intermediate Representation Language Metadata-Rich Linkable Module Format Syntax checkers for each format language listed above Software design document for system development platform Implementation of GFI-specified Phase I secure data services Phase I Challenge Problem Implementations Detailed research plan Phase I Final Report INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 46 Phase 2 HECTOR Deliverables Date Month 13 Month 14 Month 15-16 Month 18 Month 22 Month 30 Month 34 Month 36 Month 36 Event Deliverable Program Phase II Program Phase II Kick-off Meeting Third exchange on common standards Technical Exchange Meeting  TX2   Annual Site Visits  Year 2 Technical Exchange Meeting  TX3   Annual PI Program Review Meeting  Year 2   Technical Exchange Meeting  TX4   Annual PI Program Review Meeting  Year 3   Performers deliver Month-36 Deliverables and Annual Research Report System Development Platform  Version 1 Implementation of GFI-specified Phase II secure data services User guides and documentation Phase II Challenge Problem Implementations Phase II Final Report INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 47 Phase 3 HECTOR Deliverables Date Month 37 Month 39 Month 40-41 Month 44 Month 46 Month 50 Month 58 Month 60 Month 60 Event Deliverable Program Phase III Program Phase III Kick-off Meeting Forth exchange on common standards Technical Exchange Meeting  TX5   Annual Site Visits  Year 4 Technical Exchange Meeting  TX3   Annual PI Program Review Meeting  Year 4   Technical Exchange Meeting  TX4   Annual PI Program Review Meeting  Year 3   Performers deliver Month-60 Deliverables and Annual Research Report System Development Platform  Version 2 Implementation of GFI-specified Phase III secure data services Phase III Challenge Problem Implementations Phase III Final Report INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 48 Definitions of Program Metrics - Services Area Secure Data Services Metrics and Methodology Novel Secure Data Services Number of well-defined new services specified soundness verified by peer review within program P F associated overhead time number of operations Number correctly implemented correctness verified by T E P F Processing overhead Implementation of Secure Data Service Schemes INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 49 Definitions of Program Metrics - Languages Area Language Metrics and Methodology Language Specification Measure range ease of expression by making performers implement 20 unseen test problems simple routines in 1 month at end of year Use syntax checker T E evaluate ease of comprehension Measure # of correctly specified solutions # lines of code INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 50 Definitions of Program Metrics - Toolchain Toolchain Output Toolchain Metrics and Methodology Security Feasibility Study Number of security models for which accurate results are given – verified by test vectors System complexity that can be accurately handled – Number of manually verified minichallenge problems Completeness measured by mini-challenge problems Parsers syntax checkers from performers should agree Accuracy of resource estimates number of relevant quantities that can be estimated all verified manually using mini-challenge problems Efficiency of estimation Time Independent verification of formal proof of correctness of compiler functionality where provided correctness of IR output – verified using mini-challenge problems or for larger problems by observing application in action Correctness of implementation – verified by test harness live demonstration test vectors and or proof of correctness P F Auto-generated design level artifacts Intermediate Representation Resource Estimations Compiler Generated Application INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 51 Phase 1 Planned Milestones Area Phase I Entry Secure Data Services GFI Scheme list Phase I Milestones Implementation of all GFI-specified schemes selected to match challenge problems Initial baseline estimates of resource requirements Basic research plan Detailed research plan for novel secure data services concepts schemes protocols Language Toolchain System System specification language representation format specification specification document language Outline of language Implementation programming language specification including FHE SMC functional encryption GFI problem list Three GFI programming challenge problems algorithms 10 performer T E generated test routines implemented in language Ideas for IR Intermediate representation format specification Basic syntax checker design Automated syntax checkers for all new languages representation formats Linker concepts Software design concepts Linkable library module format specification Detailed software design document INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 52 Phase 2 Planned Milestones Area Phase II Entry Secure Data Services GFI Scheme list Detailed research plan Language Language and format specifications Implementation language Specifications GFI problem list Toolchain Phase II Milestones Implementation baselining of full set of interconnected secure data services Improvement of isolated protocols 10x Novel secure data services concept scheme research report how they connect with existing schemes Refined language format specifications due to implementation constraints Enhanced implementation language specification to add verifiable transparent auditable computation Full documentation for all formats and languages All GFI programming challenge problems algorithms and performer T E generated routines implemented Syntax checkers Compilers and other toolchain artifacts implemented Linkable module format Linkable modules derived from secure data services Software design document Demonstration of toolchain compiling verifying Phase 1 GFI challenge problems INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 53 Phase 3 Planned Milestones Area Phase III Entry Phase III Milestones Secure Data Services Implementation of optimized Demonstrated efficiency improvements 10x in GFI-specified schemes interconnected schemes 100x in isolated schemes All new performer concepts Implementation of secure data service concepts schemes schemes protocols from own and or other performer research Language Refined language format specifications Enhanced implementation language to add at least one new concept from program research results Full documentation Full language documentation including at least one new concept All GFI challenge problems 3 selected additional performer-suggested implemented programming challenge problems algorithms implemented in language Toolchain Toolchain implemented Implementation of at least one new language concept in toolchain Optimization of toolchain performance 2x Implementation of optimizing compiler within toolchain INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 54 Measurement of Success Challenge Problems Test Evaluation GFI challenges from real-world problems Use existing metrics techniques for Can exercise entire toolchain See implementation of new concepts “Chinese menu” maximizes breadth Language acceptability criteria Assessing representation formats Testing complier toolchains Performer generated challenge problems Toolchain output allows review of Demonstrate range of expression Require “future-proof” software Automated security analyses Resource estimation capabilities Pick Your Technologies Pick a Problem Pick the Adversaries Functional Encryption Census Data Processing Malicious Verifiable Computation Health Record Processing Honest But Curious Oblivious RAM Sealed Auctions Covert Malicious INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 55 Program Roles and Responsibilities  Performers  Research Development  Government Support  Government Furnished Information GFI  Challenge problems  At the kickoff of each program phase the Government will provide performers with a list of benchmarks and background information as GFI  Government Furnished Equipment GFE  None  Testing and Evaluation  Metrics for evaluation  Yearly report on performer progress so far  Quarterly cross-performer meetings per focus area INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 56 Government Furnished Information GFI  The followings are examples of GFI  Secure data services to be implemented by performers including full specification or reference to same  Specification of micro-scale challenge problems to be used to demonstrate the correct operation of tools  Specification of large-scale challenge problems to be used to demonstrate the application of the system development platform to realworld problems INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 57 Challenge Problems  The HECTOR program will use a variety of challenge problems to allow performers to demonstrate the merits of their programming languages and representation formats and to demonstrate the successful operation of their system development platform  Challenge problems will be provided as Government Furnished Information GFI at the outset of each phase Initial challenge problems will be provided at program kickoff Example Applications Sealed Auction Treaty Negotiation Policy-compliant data processing Outsourced data processing Secure Election INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 58 Challenge Problems  Challenge problems will be given in two varieties small-scale challenges such as invoking a specific secure data service and system-scale challenges such as implementing an entire secure data processing system  The system-scale challenges will be picked from a “menu” of possibilities through which multiple paths can be generated by selecting different attributes at each stage of the menu  A preliminary list of menu items is provided in the next chart INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 59 Challenge Problems – Menu Items Attribute Number of Participants Participant Dynamism Adversarial model s Network Latency Protection of inputs Use of outputs Data Services Required Description How many total nodes  and or users will be  parties in the system Possible Values 2 10 100 1000 Whether parties in the  Fixed system are fixed or  Some changes at run-time Ad-hoc dynamic The models of possible  Benign adversarial behavior to be  Honest but curious assumed at specific nodes  Covert Malicious and or links between  Custom nodes The latency of specific  High  e g  satellite link Medium  e g  transcontinental link links within the system  Low  e g  local link architecture How input data are  Visible to members only protected from the  Also visible to the compute engine compute system or from  Private from others Verifiable properties external entities No security properties How data exits the system Direct decrypt Encrypted for later reuse Encrypted for specific party Proxy re-encryption One or more capabilities  Auditable computation to be implemented Verifiable computation Compulsory policy compliance Secure multiparty computation Functional encryption Multiparty Signoff Homomorphic encryption INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA T E Team Roles and Responsibilities  Multidisciplinary T E team draws from multiple sources     SETA oversight and reporting FFRDC cryptographic expertise Academia compiler design Industry large scale software design system specifications INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 61 Risks • Excessive Overhead – We do not know whether the known overheads of advanced cryptographic techniques will still be excessively onerous when incorporated into the system as a whole – Mitigation if HECTOR cannot bring those overheads down it can at least identify bottlenecks at the system level and show what could be achieved with a specific further reduction • Automated Security Analysis – While the system specification format should ease the process flexible representation of system goals and security threats to generate a thorough and meaningful automated security analysis is currently an unsolved problem – Mitigation by forcing up-front declaration of these system properties HECTOR will at least ease any subsequent manual analysis INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 62 HECTOR Timeline Project start Phase 3 Optimize Refine Phase 2 Implement Phase 1 Plan Design Month 12 • Develop detailed plans designs and specifications and build automated grammar and syntax checkers • Demonstrate viability of plans and system designs Month 36 • Research new concepts for secure data services performance improvements in existing services • Implement additional secure data services • Implement all system development platform functionality and tools • Develop exemplar applications to address the challenge problems Month 60 • Research implement performance improvements and additional secure data services • Incorporate optimization strategies • Improve the accuracy performance of resource estimation tools • Develop exemplar applications to address the challenge problems INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 63 Reporting Requirements • Monthly technical report – highlight progress from past month and plans for next month • Monthly financial report – form will be provided • Program kick-off meeting – first month of program • Annual performer site visit – beginning of each program year • Technical Exchange Meetings • Semi-annual Program Review Meetings • Test Plans • Reports – submitted at the end of each year INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 64 Notional Target Schedule FY17 FY18 FY19 FY20 FY21 FY22 FY23 Proposer’s Day BAA Phase 1 Phase 2 Phase 1 Schedule Proposer’s Day BAA Release TBR Proposals Due TBR Source Selection TBR Program Kickoff TBR Phase 3 July 26 September 15 October 30 December 31 April 15 BAA Review and Source Selection INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 65 Management Plan and Teaming • Depth and diversity will be essential to accomplish the many challenges in tool development and extension • Scalability and Optimization • Make sure you have enough people both from industry and academia to accomplish the goal and from proof-of-concept to large scale • Sufficient resources to follow critical path while still exploring new approaches • Completeness – teams should not lack any capability necessary for success e g should not rely upon results or enabling technology from the community at large • Tightly knit teams • • Clear strong management single point of contact • No loose confederations No teaming for teaming’s sake • Each team member should contribute significantly to the program goals Team members not required to participate all 5 years – consider phase transitions INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 66 Proposal Evaluation Criteria • • Evaluation criteria in descending order of importance are • Overall technical merit • Effectiveness of proposed work plan • Relevance to IARPA mission and HECTOR program goals • Relevant experience and expertise of the members of the team • Cost realism All responsive proposals will be evaluated by a board of qualified government reviewers INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 67 Point of Contact Dr Mark I Heiligman Program Manager IARPA Office of the Director of National Intelligence Intelligence Advanced Research Projects Activity Washington DC 20511 Phone 301 851-7432 Fax 301 851-7672 Electronic mail dni-iarpa-baa-17-05@iarpa gov include IARPA-BAA-17-05 in the Subject Line Website www iarpa gov Questions Please fill out cards INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 68 DFFIGE THE DIRECTOR NATIONAL INTELLIGENCE Questions Eligibility Information • Collaborative efforts are strongly encouraged • • Content communications networking and team formation is the responsibility of proposers Foreign organizations and or individuals are welcome to participate • Must comply with Non-Disclosure Agreements Security Regulations Export Control Laws etc as appropriate • Other Government Agencies Federally Funded Research and Development Centers FFRDCs University Affiliated Research Centers UARCs and any organizations that have a special relationship with the Government including access to privileged and or proprietary information or access to Government equipment or real property are not eligible to submit proposals under this BAA or participate as team members under proposals submitted by eligible entities • Please notify the HECTOR Program Manager ASAP if you wish to utilize any resources from these organizations • If IARPA determines that the resources are unique and do not exist in the private sector IARPA will attempt to work directly with that organization to arrange for that capability to be made available to all program participants who might benefit INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 70 Doing Business with IARPA Mark Heiligman Intelligence Advanced Research Projects Activity INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA HECTOR Proposers’ Day Agenda Time Topic 9 00 am – 9 30 am Registration and Check In 9 30 am – 9 45 am IARPA Overview and Remarks 9 45 am – 10 30 am HECTOR Program Overview 10 30 am – 11 00 am BAA Overview T E GFI GFE 11 00 am – 11 30 am Break 11 30 am – 12 00 pm Doing Business with IARPA 12 00 pm – 12 30 pm HECTOR Program Questions Answers 12 30 pm – 1 30 pm Lunch 1 30 pm – 3 00 pm Proposers' 5-minute Capability Presentations 3 00 pm – 4 00 pm Proposers’ Networking and Teaming Discussions Speaker IARPA management Mark Heiligman Program Manager Mark Heiligman Program Manager IARPA Acquisition Mark Heiligman Program Manager Attendees No Government Attendees No Government INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA Doing Business with IARPA - Recurring Questions Questions and Answers http www iarpa gov index php faqs Eligibility Info Intellectual Property Pre-Publication Review Preparing the Proposal Broad Agency Announcement BAA Section 4 Electronic Proposal Delivery https iarpa-ideas gov • Organizational Conflicts of Interest http www iarpa gov index php working-with-iarpa iarpas-approach-to-oci Streamlining the Award Process Accounting system Key Personnel IARPA Funds Applied Research RECOMMENDATION Please read the entire BAA INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 73 Responding to Q As Please read entire BAA before submitting questions Pay attention to Section 4 Proposal Submission Information Read Frequently Asked Questions on the IARPA @ http www iarpa gov index php faqs Send your questions as soon as possible HECTOR BAA dni-iarpa-baa-17-05@iarpa gov Write questions as clearly as possible Do NOT include proprietary information INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 74 Eligible Applicants Collaborative efforts teaming strongly encouraged Content communications networking and team formation are the responsibility of Proposers Foreign organizations and or individuals may participate Must comply with Non-Disclosure Agreements Security Regulations Export Control Laws etc as appropriate as identified in the BAA INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 75 Ineligible Organizations Other Government Agencies Federally Funded Research and Development Centers FFRDCs University Affiliated Research Centers UARCs and any organizations that have a special relationship with the Government including access to privileged and or proprietary information or access to Government equipment or real property are not eligible to submit proposals under this BAA or participate as team members under proposals submitted by eligible entities INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 76 Intellectual Property IP Unless otherwise requested Government rights for data first produced under IARPA contracts will be UNLIMITED At a minimum IARPA requires Government Purpose Rights GPR for data developed with mixed funding Exception to GPR State in the proposal any restrictions on deliverables relating to existing materials data software tools etc INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 77 Pre-Publication Review Funded Applied Research efforts IARPA encourages Publication for Peer Review of UNCLASSIFIED research Prior to public release of any work submitted for publication the Performer will Provide copies to the IARPA PM and Contracting Officer Representative COR COTR Ensure shared understanding of applied research implications between IARPA and Performers IARPA PM decides on approval for release or receiving courtesy copy INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 78 Preparing the Proposal Note restrictions in BAA Section 4 on proposal submissions Interested Offerors must register electronically IAW instructions on https iarpa-ideas gov Interested Offerors are strongly encouraged to register in IDEAS at least 1 week prior to proposal “Due Date” Offerors must ensure the version submitted to IDEAS is the “Final Version” Classified proposals – Contact IARPA Chief of Security BAA format is established to answer most questions Check FBO for amendments IARPA website for Q As BAA Section 5 – Read Evaluation Criteria carefully e g “The technical approach is credible and includes a clear assessment of primary risks and a means to address them” INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 79 Preparing the Proposal BAA Sect 4 Read IARPA’s Organizational Conflict of Interest OCI policy http www iarpa gov index php working-with-iarpa iarpas-approach-to-oci See also eligibility restrictions on use of Federally Funded Research and Development Centers University Affiliated Research Centers and other similar organizations that have a special relationship with the Government Focus on possible OCIs of your institution as well as the personnel and subcontractors on your team See Section 4 It specifies the non-Government e g SETA FFRDC UARC etc support we will be using If you have a potential or perceived conflict request a waiver as soon as possible INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 80 Organizational Conflict of Interest OCI If a prospective offeror or any of its proposed subcontractor teammates believes that a potential conflict of interest exists or may exist whether organizational or otherwise the offeror should promptly raise the issue with IARPA and submit a waiver request by e-mail to the mailbox address for this BAA at dni-iarpa-baa-17-05@iarpa gov A potential conflict of interest includes but is not limited to any instance where an offeror or any of its proposed subcontractor teammates is providing either scientific engineering and technical assistance SETA or technical consultation to IARPA In all cases the offeror shall identify the contract under which the SETA or consultant support is being provided Without a waiver from the IARPA Director neither an offeror nor its proposed subcontractor teammates can simultaneously provide SETA support or technical consultation to IARPA and compete or perform as a Performer under this solicitation INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 81 Streamlining the Award Process Cost Proposal – we only need what we ask for in BAA Approved accounting system needed for Cost Reimbursable contracts Must be able to accumulate costs on job-order basis DCAA or cognizant auditor must approve system See http www dcaa mil “Audit Process Overview - Information for Contractors” under the “Guidance” tab Statements of Work format may need to be revised Key Personnel Expectations of time note the Evaluation Criteria requiring relevant experience and expertise Following selection Contracting Officer may request your review of subcontractor proposals INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 82 IARPA Funding IARPA funds Applied Research for the Intelligence Community IC IARPA cannot waive the requirements of Export Administrative Regulation EAR or International Traffic in Arms Regulation ITAR Not subject to DoD funding restrictions for R D related to overhead rates IARPA is not DoD INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 83 Disclaimer This is Applied Research for the Intelligence Community Content of the Final BAA will be specific to this program The Final BAA is being developed Following issuance look for Amendments and Q As There will likely be changes The information conveyed in this brief and discussion is for planning purposes and is subject to change prior to the release of the Final BAA INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 84 Point of Contact Dr Mark I Heiligman Program Manager IARPA Office of the Director of National Intelligence Intelligence Advanced Research Projects Activity Washington DC 20511 Phone 301 851-7432 Fax 301 851-7672 Electronic mail dni-iarpa-baa-17-05@iarpa gov include IARPA-BAA-17-05 in the Subject Line Website www iarpa gov Questions Please fill out cards INTELLIGENCE ADVANCED RESEARCH PROJECTS ACTIVITY IARPA 85