Privacy Preserving Computations (PPC) that utilize Homomorphic Encryption (HE) to perform <br/>computations on encrypted data directly have become attractive recently. HE based Machine Learning (HE <br/>ML) inference enables preservation of privacy in a wide variety of application domains that range from <br/>healthcare, financial transactions, edge cyber physical systems, etc. Privacy sensitive applications that rely <br/>on processing in a public cloud or data center can use HE ML inference to preserve privacy. While HE ML <br/>inference offers strong privacy guarantees, computations on encrypted data are orders of magnitude slower <br/>than unencrypted computations and require significant hardware resources to make them attractive for end <br/>users. Emerging data centers and cloud platforms are augmented with Field Programmable Gate Arrays <br/>(FPGAs). With the fine grained programmable architecture of FPGAs, these platforms are well suited for <br/>accelerating HE ML.<br/><br/>This work will leverage novel algorithmic, architectural and memory optimizations on FPGAs to develop a <br/>portable and configurable library to enable secure, resilient, and trustworthy cyberinfrastructure for end-to-<br/>end privacy sensitive ML inference. The library will provide FPGA accelerated Intellectual Property (IP) <br/>cores for HE kernels (L1 Library) as well as a FPGA Application Specific Processor (ASP) for inference of <br/>widely studied HE ML models (L2 Library). The library will support various HE schemes, security levels, <br/>machine learning models and FPGA platforms. It will include several software and hardware innovations <br/>along with various HE specific optimizations such as efficient data layout, memory efficient scheduling and <br/>scalable interconnect to maximize memory utilization and to improve data reuse using on-chip memory. <br/>Using the IP cores in the L1 Library, this project will compose a FPGA ASP with a domain-specific <br/>Instruction Set Architecture (ISA) and a compiler. The FPGA accelerator can be programmed in software to <br/>realize real-time HE ML computations. The library will be released to the Computer & Information Science <br/>& Engineering (CISE) communities, including Machine Learning, Software, and Data Science communities, <br/>to accelerate the adoption of homomorphic encryption for privacy preserving computations.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.