ICDCN 2024 Keynote Speakers
School of Engineering and Applied Sciences, Harvard University.
Title: Reasoning on Uncertain Learned Data
Brief Abstract: This basic question faces humans in everyday experience. It is also an important technological problem if we are to make good use of the power of machine learning. In this talk this question will be discussed in the framework of a new theory of human capabilities for information processing.
Bio: Leslie Valiant was educated at King's College, Cambridge, Imperial College London, and the University of Warwick, where he received a PhD in computer science in 1974.
He won the Turing Award in 2010 with a citation that reads, "For transformative contributions to the theory of computation, including the theory of probably approximately correct (PAC) learning, the complexity of enumeration and of algebraic computation, and the theory of parallel and distributed computing."
Other accolades include the Nevanlinna Prize in 1986, the Knuth Prize in 1997, and the EATCS Award in 2008. He was elected a Fellow of the Royal Society (FRS) in 1991, a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI) in 1992, and a member of the United States National Academy of Sciences in 2001.
Professor and Chair, Computer Science
Stony Brook University - SUNY
Title: Passive Backscattering RF Tag Networks - Fundamentals and Applications
Abstract: We consider a new generation of Internet-of-Things (IoT) technology that uses 'passive' backscattering RF tags operating in an extremely low-power regime. The most common use of such tags is in RFID technology, where a separate active radio transceiver ('reader') is used to communicate with the passive tags. However, this limits the scalability and broad applicability of such tags. Our work considers recent advances where such passive tags can communicate among themselves without any active radio device. In this talk, we will describe the fundamental technology of such tag-to-tag communication, including techniques to improve the range and robustness of the tag-to-tag links. We will also describe a novel RF-based sensing technique to estimate the tag-to-tag channel parameters and showcase how such sensing can enable wide-ranging applications including localization and monitoring.
Professor, Computer Science, MIT
Title: Theoretical Foundations of Real-World Cryptography
Abstract: Block ciphers (aka Pseudorandom Permutations) are the workhorses of modern cryptography. An overwhelming fraction of Internet traffic is secured using the Advanced Encryption Standard (AES), a block cipher; the same goes for encrypted file systems, encrypted messaging, and several other applications we use all the time. In this talk, I will describe a new research program, together with a collection of recent results, aimed at achieving a foundational mathematical understanding of block ciphers.
Co-founder & Co-CEO
Title: Hashgraph Consensus
Abstract: The hashgraph consensus algorithm allows multiple computers to reach agreement with finality in a way that is very fast and very secure (Asynchronous Byzantine Fault Tolerant: ABFT). We will talk about some of the ideas behind it, how it works, how this affects blockchain and Distributed Ledger Technology (DLT), and some of the areas where research is ongoing.
Professor, Computer Science
Title: One-Way Functions Candidates Based on Information Theoretical Secure Primitives
Abstract: The talk consists of recently proposed one-way functions (OWF) candidates that employ perfect and provable information theoretical secure techniques, for example, one-time pad and secret sharing. The potential of leveraging the unique properties of, secret sharing, one-time pad, and random permutations in the design of effective one-way functions is the exploration motivation.
For example, one such design involves applying the exclusive-or (xor) operation to two randomly chosen strings. To address concerns related to preimage mappings, we incorporate error detection codes. Additionally, we utilize permutations to overcome linearity issues in the computation process. In order to enhance the security of our approach, we propose the integration of a secret-sharing scheme based on a linear polynomial. This helps mitigate collisions and adds an additional layer of perfect security. We thoroughly investigate the interactions between different aspects of one-way functions to strengthen the reliability of commitments. Lastly, we explore the possibility of nesting one-way functions as a countermeasure against potential backdoors. Through our study, we aim to contribute to the advancement of secure encryption techniques by leveraging the inherent strengths of the one-time pad and carefully considering the interplay of various components in the design of one-way functions. The talk is based on recent works in collaboration with Asaf Choen, Pawel Cyprys, Hagar Dolev, Oded Margalit, and Shlomo Moran.