London-ish Lattice & Coding Meeting: 18 January 2017

[Martin R.Albrecht]

Dear all,

Lattice-based approaches are emerging as a common theme in modern
cryptography and coding theory. In communications, they are
indispensable mathematical tools to construct powerful error-correction
codes achieving the capacity of wireless channels. In cryptography, they
are used to building lattice-based schemes with provable security,
better asymptotic efficiency, resilience against quantum attacks and new
functionalities such as fully homomorphic encryption.

This meeting — on *18 January 2017* — is aimed at connecting the two
communities in the UK with a common interest in lattices, with a
long-term goal of building a synergy of the two fields. It will consist
of several talks on related topics, with a format that will hopefully
encourage interaction.

# Tentative program #

More details here as they become available:

http://malb.io/discrete-subgroup/2017/01/18/

## Wouter Castryck: Remarks on the error distributions in ring-based LWE ##

The existing literature contains several ring-based variants of the Learning With Errors problem, all of which are often referred to as Ring-LWE, a habit which has led to some confusion in the recent past. The main difference lies in the choice of the probability distribution from which the errors are to be sampled. We will briefly compare the main versions, such as Poly-LWE and "proper" Ring-LWE as introduced by Lyubashevsky et al., and discuss some pitfalls that arise when mixing things up. This is joint work with Ilia Iliashenko and Frederik Vercauteren.

## Ana Costache:</span> Fixed Point Arithmetic in SHE Scheme ##

## Victor Beresnevich: TBC ##

## Anne-Maria Ernvall-Hytönen: TBC ##

# Venue #

Room 611 (Dennis Gabor Seminar Room)
Department of Electrical and Electronic Engineering
Imperial College London
South Kensington
London SW7 2AZ

## Directions ##

http://www.imperial.ac.uk/visit/campuses/south-kensington/

# Registration #

Everyone is welcome. Two caveats:

1. Speakers are told the audience is somewhat familiar with lattices.

2. Please send us an email at <c.l...@imperial.ac.uk>, so that the size
of the room fits with the number of participants.

Best,
Cong Ling and Martin Albrecht

--

_pgp: https://keybase.io/martinralbrecht
_www: https://martinralbrecht.wordpress.com
_jab: martinr...@jabber.ccc.de
_otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF

[Martin R.Albrecht]

Hi all,

the next meeting is now less than two weeks away. Below is the updated
programme. Hope to see many of you there.

# Tentative program #

More details here as they become available:

http://malb.io/discrete-subgroup/2017/01/18/

## Wouter Castryck: Remarks on the error distributions in ring-based LWE ##

The existing literature contains several ring-based variants of the Learning With Errors problem, all of which are often referred to as Ring-LWE, a habit which has led to some confusion in the recent past. The main difference lies in the choice of the probability distribution from which the errors are to be sampled. We will briefly compare the main versions, such as Poly-LWE and "proper" Ring-LWE as introduced by Lyubashevsky et al., and discuss some pitfalls that arise when mixing things up. This is joint work with Ilia Iliashenko and Frederik Vercauteren.

## Ana Costache:</span> Fixed Point Arithmetic in SHE Scheme ##

We begin by introducing the context of ring-based somewhat homomorphic schemes and discuss some optimisations. This starts with introducing the RLWR hard problem and its relation to lattices. We investigate fixed-point arithmetic in ring-based homomorphic encryption schemes. Downlin et al. present two fixed-point numbers representations; we analyse and show them to be isomorphic, by presenting an explicit isomorphism between the two. Given input bounds on fixed-point numbers and scalars, we achieve lower bounds for the ring dimensions needed to support complex homomorphic operations. As an application, we investigate homomorphic image processing and, specifically, Fourier Transforms.

## Victor Beresnevich: TBC ##

## Anne-Maria Ernvall-Hytönen: Secrecy Function and Comparing Lattices ##

In this talk, I will first explain how theta functions of lattices appear in analysing which lattices are good for coset coding in wiretap channels. I will then explain what ia the secrecy gain of a lattice, and what is the conjecture of Belfiore and Sole for unimodular lattices, and how it has been generalised for l-modular lattices. I will then explain how this conjecture can be approached for both in the unimodular case, and for various values of l. Finally, I will explain about the problems in using secrecy gains in comparing lattices.