In September 2022, J.P. Morgan launched the firm-wide AIgoCRYPT Center of Excellence (CoE) to lead cutting-edge research in cryptography and secure distributed (AI) computation. The AIgoCrypt CoE is a cross functional team of researchers from AI Research and partners across the business whose objective is to design, share and implement state-of-the-art techniques which enable secure computation of encrypted data. The members of the CoE maintain a close connection with academia and the scientific field by publishing in top cryptography/security venues, participating in scientific events, disseminating knowledge and collaborating with other researchers around the globe.

The center advances the state-of-the-art in multiple directions, including:

  • Secure multiparty computation (MPC)
  • Fully homomorphic encryption (FHE)
  • Privacy-preserving machine learning (PPML)
  • Privacy-preserving federated learning (PPFL)
  • Zero-knowledge proofs (ZK)
  • Differential privacy (DP)
  • Quantum cryptography (QC)
  • Privacy-preserving blockchains

List of publications

Lightweight, Maliciously Secure Verifiable Function Secret Sharing. In EUROCRYPT’22
Leo de Castro and Antigoni Polychroniadou

Kicking-the-Bucket: Fast Privacy-Preserving Trading Using Buckets
Mariana Botelho da Gama, John Cartlidge, Antigoni Polychroniadou, Nigel P. Smart and Younes Talibi Alaoui

Sharing Transformation and Dishonest Majority MPC with Packed Secret Sharing. In CRYPTO'22
Vipul Goyal, Antigoni Polychroniadou and Yifan Song

More Efficient Dishonest Majority Secure Computation over Z_2^k via Galois Rings. In CRYPTO'22
Daniel Escudero, Chaoping Xing and Chen Yuan

Fast Fully Secure Multi-Party Computation over Any Ring with Two-Thirds Honest Majority. In CCS'22 (The ACM Conference on Computer and Communication Security)
Anders Dalskov, Ariel Nof and Daniel Escudero

TurboPack: Honest Majority MPC with Constant Online Communication. In CCS'22 (The ACM Conference on Computer and Communication Security)
Daniel Escudero, Vipul Goyal, Antigoni Polychroniadou and Yifan Song

Vector Commitments over Rings and Compressed Sigma-Protocols. In TCC'22 (Theory of Cryptography Conference)
Thomas Attema, Ignacio Cascudo, Ronald Cramer, Ivan Bjerre Damgård and Daniel Escudero

 

ACCO: Algebraic Computation with Comparison. In CCSW’21 (Cloud Computing Security Workshop)
Xiaoqi Duan, Vipul Goyal, Hanjun Li, Rafail Ostrovsky, Antigoni Polychroniadou and Yifan Song

Round-Optimal Secure Multi-party Computation. In Journal of Cryptology 2021
Shai Halevi, Carmit Hazay, Antigoni Polychroniadou, Muthuramakrishnan Venkitasubramaniam

Privacy Preserving Portfolio Pricing. In ICAIF’21 (International Conference on AI in Finance)
Gilad Asharov, Tucker Balch and Antigoni Polychroniadou

Improved Single-Round Secure Multiplication Using Regenerating Codes. In ASIACRYPT’21
Mark Abspoel, Ronald Cramer, Daniel Escudero, Ivan Damgård and Chaoping Xing

Information-Theoretically Secure MPC against Mixed Dynamic Adversaries. In TCC’21
Ivan Damgård, Daniel Escudero and Divya Ravi

Unconditional Communication-Efficient MPC via Hall’s Marriage Theorem. In 41st Annual CRYPTO’21
Vipul Goyal, Yifan Song and Antigoni Polychroniadou

ATLAS: Efficient and Scalable MPC in the Honest Majority Setting. In 41st Annual CRYPTO'21
Rafail Ostrovsky, Antigoni Polychroniadou, Yifan Song, Vipul Goyal and Hanjun Li

Constant-Overhead Unconditionally Secure Multiparty Computation over Binary Fields. In 40th Annual EUROCRYPT’21
Yifan Song and Antigoni Polychroniadou

 

Succinct Non-interactive Secure Computation. In EUROCRYPT’20.
Andrew Morgan, Rafael Pass and Antigoni Polychroniadou

Privacy-Preserving Dark Pools. In AAMAS’20 (International Conference on Autonomous Agents and Multiagent Systems)
Gilad AsharovTucker Balch, Antigoni Polychroniadou and Manuela Veloso

Small Memory Robust Simulation of Client-Server Interactive Protocols over Oblivious Noisy Channels. In SODA’20
T.-H. Hubert Chan, Zhibin Liang, Antigoni Polychroniadou and Elaine Shi

SecretMatch: Inventory Matching from Fully Homomorphic Encryption. In ICAIF’20
Tucker BalchBenjamin Diamond and Antigoni Polychroniadou

CryptoCredit: Securely Training Fair Models. In ICAIF’20 (International Conference on AI in Finance)
Leo de Castro, Jiahao Chen and Antigoni Polychroniadou

Differentially Private Secure Multi-Party Computation for Federated Learning in Financial Applications. In ICAIF’20
David Byrd and Antigoni Polychroniadou

 

Manuscripts

Paving the Way towards 800 Gbps Quantum-Secured Optical Channel Deployment in Mission-Critical Environments.
Farzam Toudeh-Fallah, Marco Pistoia, Yasushi Kawakura, Navid Moazzami, David H. Kramer, Robert I. Woodward, Greg Sysak, Benny John, Omar Amer, Antigoni O. Polychroniadou, Jeffrey Lyon, Suresh Shetty, Tulasi D. Movva, Sudhir Upadhyay, Monik R. Behera, Joseph A. Dolphin, Paul A. Haigh, James F. Dynes and Andrew J. Shields

Prio+: Privacy Preserving Aggregate Statistics via Boolean Shares.
Surya Addanki, Kevin Garbe, Eli Jaffe, Rafail Ostrovsky and Antigoni Polychroniadou

Phoenix: Secure Computation in an Unstable Network with Dropouts and Comebacks.
Ivan Damgård, Daniel Escudero and Antigoni Polychroniadou

 

This [paper/presentation] was prepared for informational purposes by the [Artificial Intelligence Research] group of JPMorgan Chase & Co and its affiliates (“JP Morgan”), and is not a product of the Research Department of JP Morgan. JP Morgan makes no representation and warranty whatsoever and disclaims all liability, for the completeness, accuracy or reliability of the information contained herein. This document is not intended as investment research or investment advice, or a recommendation, offer or solicitation for the purchase or sale of any security, financial instrument, financial product or service, or to be used in any way for evaluating the merits of participating in any transaction, and shall not constitute a solicitation under any jurisdiction or to any person, if such solicitation under such jurisdiction or to such person would be unlawful.