Integrated Calendar

TBD

A distribution is k-incompressible, Yao [FOCS ’82], if no efficient compression scheme compresses it to less than k bits. While being a natural measure, its relation to other computational analogs of entropy such as pseudoentropy (Hastad, Impagliazzo, Levin, and Luby [SICOMP 99]), and to other cryptographic hardness assumptions, was unclear.

We advance towards a better understating of this notion, showing that a k-incompressible distribution has (k-2) bits of next-block pseudoentropy, a refinement of pseudoentropy introduced by Haitner, Reingold, and Vadhan [SICOMP ’13]. We deduce that a samplable distribution X that is (H(X) 2)-incompressible, implies the existence of one-way functions.

Joint work with Iftach Haitner and Jad Silbak.

Climate change is having an impact on agricultural production and food
security. Rising temperatures, changes in rainfall patterns and extreme
weather events can reduce crop yields, sometimes dramatically. However,
climate change can also offer new opportunities, by generating more
favorable climatic conditions for agricultural production in certain regions
that were previously less productive. In order to assess the positive and
negative impacts of climate change on agriculture and identify effective
adaptation strategies, scientists have produced massive amounts of data
during the last two decades, conducting local experiments in agricultural
plots and using models to simulate the effect of climate on crop yields. In
most cases, these data are not pooled together and are analyzed separately
by different groups of scientists to assess the effects of climate change at a
local level, without any attempt to upscale the results at a larger scale. Yet, if
brought together, these data represent a rich source of information that are
relevant to analyze the effect of climate across diverse environmental
conditions. The wealth of data available has led to the emergence of a new
type of scientific activity, involving the retrieval of all available data on a
given subject and their synthesis into more robust and generic results. In this
talk, I review the statistical methods available to synthesize data generated
in studies quantifying the effect of climate change on agriculture. I discuss
both the most classic methods - such as meta-analysis - and more recent
methods based on machine learning. In particular, I show how this approach
can be used to map the impact of climate change on a large scale (national,
continental and global) from local data. I illustrate these methods in several
case studies and present several research perspectives in this area.