A deterministic extractor for an elliptic curve is a function that converts a random point on the curve to a random-looking bit-string, which is statistically close to a uniformly random bit-string. The problem of converting random points of an elliptic curve into random bits has several cryptographic applications. In this talk, we propose two simple and efficient deterministic extractors for a...

A recent Ontario arbitral award is reshaping the way employers and employees must approach the legal duty to accommodate employees with disabilities in the workplace. Entrenched in both federal and provincial human rights legislation, the duty to accommodate imposes a legal obligation on employers to accommodate employees with disabilities unless such accommodation would cause undue hardship on...

We analyze the distribution of ∑ m i=1 vixi where x1, . . . ,xm are fixed vectors from some lattice L ⊂ R (say Z) and v1, . . . , vm are chosen independently from a discrete Gaussian distribution over Z. We show that under a natural constraint on x1, . . . ,xm, if the vi are chosen from a wide enough Gaussian, the sum is statistically close to a discrete Gaussian over L. We also analyze the cas...

We propose two simple and efficient deterministic extractors for J(Fq), the Jacobian of a genus 2 hyperelliptic curve H defined over Fq, for some odd q. Our first extractor, SEJ, called sum extractor, for a given point D on J(Fq), outputs the sum of abscissas of rational points on H in the support of D, considering D as a reduced divisor. Similarly the second extractor, PEJ, called product extr...

Given two distributions over an n element set, we wish to check whether these distributions are statistically close by only sampling. We give a sublinear algorithm which uses O(n2/3ǫ−4 logn) independent samples from each distribution, runs in time linear in the sample size, makes no assumptions about the structure of the distributions, and distinguishes the cases when the distance between the d...

It is known that if a 2-universal hash function H is applied to elements of a block source (X1, . . . , XT ), where each item Xi has enough min-entropy conditioned on the previous items, then the output distribution (H,H(X1), . . . , H(XT )) will be “close” to the uniform distribution. We provide improved bounds on how much min-entropy per item is required for this to hold, both when we ask tha...

The classic Leftover Hash Lemma (LHL) is often used to argue that certain distributions arising from modular subset-sums are close to uniform over their finite domain. Though very powerful, the applicability of the leftover hash lemma to lattice based cryptography is limited for two reasons. First, typically the distributions we care about in lattice-based cryptography are discrete Gaussians, n...

In this paper, we propose a new score normalization method for text-independent speaker verification using GMM (Gaussian Mixture Model). In the proposed method, cohort model is designed as virtual speaker model based on the similarity of local acoustic information between the reference speaker and other customers. The similarity is determined using statistical distance between model components ...

The ideal choices for the tasks presented in the project proposal would be the NVIDIA CUDA toolkit (http://developer.nvidia.com/object/cuda. html), which exposes more underlying architecture to programmers. However, the package requires a capable NVIDIA video card, and we could not get for this project. ATI also designed a similar platform “Close-to-Metal (CTM) Device” (http://ati.de/companyinf...