“Market-Basket (MB) and Household (HH) data provide a fertile substrate for the inference of association between marketing activity (e.g.: prices, promotions, advertisement, etc.) and customer behaviour (e.g.: customers driven to a store, specific product purchases, joint product purchases, etc.). The main aspect of MB and HH data which makes them suitable for this type of inference is the large number of variables of interest they contain at a granularity that is fit for purpose (e.g.: which items are bought together, at what frequency are items bought by a specific household, etc.).

A large number of methods are available to researchers and practitioners to infer meaningful networks of associations between variables of interest (e.g.: Bayesian networks, association rules, etc.). Inferred associations arise from applying statistical inference to the data. In order to use statistical association (correlation) to support an inference of causal association (“which is driving which”), an explicit theory of causality is needed.

Such a theory of causality can be used to design experiments and analyse the resultant data; in such a context certain statistical associations can be interpreted as evidence of causal associations.

On observational data (as opposed to experimental), the link between statistical and causal associations is less straightforward and it requires a theory of causality which is formal enough to support an appropriate calculus (e.g.: do-calculus) of counterfactuals and networks of causation.

My talk will be focused on providing retail analytic problems which may motivate an interest in exploring causal calculi’s potential benefits and challenges.”

To face the advent of multicore processors and the ever increasing complexity of hardware architectures, programming
models based on DAG parallelism regained popularity in the high performance, scientific computing community. Modern runtime systems offer a programming interface that complies with this paradigm and powerful engines for scheduling the tasks into which the application is decomposed. These tools have already proved their effectiveness on a number of dense linear algebra applications. 

In this talk we present the design of task-based sparse direct solvers on top of runtime systems. In the context of the
qr_mumps solver, we prove the usability and effectiveness of our approach with the implementation of a sparse matrix multifrontal factorization based on a Sequential Task flow parallel programming model. Using this programming model, we developed features such as the integration of dense 2D Communication Avoiding algorithms in the multifrontal method allowing for better scalability compared to the original approach used in qr_mumps.

Following this approach, we move to heterogeneous architectures where task granularity and scheduling strategies are critical to achieve performance. We present, for the multifrontal method, a hierarchical strategy for data partitioning and a scheduling algorithm capable of handling the heterogeneity of resources.   Finally we introduce a memory-aware algorithm to control the memory behavior of our solver and show, in the context of multicore architectures, an important reduction of the memory footprint for the multifrontal QR factorization with a small impact on performance.

In the year 2003 Yuri Ershov gave a talk at a conference in Teheran on
his notion of ``extremal valued fields''. He proved that algebraically
complete discretely valued fields are extremal. However, the proof
contained a mistake, and it turned out in 2009 through an observation by
Sergej Starchenko that Ershov's original definition leads to all
extremal fields being algebraically closed. In joint work with Salih
Durhan (formerly Azgin) and Florian Pop, we chose a more appropriate
definition and then characterized extremal valued fields in several
important cases.

We call a valued field (K,v) extremal if for all natural numbers n and
all polynomials f in K[X_1,...,X_n], the set of values {vf(a_1,...,a_n)
| a_1,...,a_n in the valuation ring} has a maximum (which is allowed to
be infinity, attained if f has a zero in the valuation ring). This is
such a natural property of valued fields that it is in fact surprising
that it has apparently not been studied much earlier. It is also an
important property because Ershov's original statement is true under the
revised definition, which implies that in particular all Laurent Series
Fields over finite fields are extremal. As it is a deep open problem
whether these fields have a decidable elementary theory and as we are
therefore looking for complete recursive axiomatizations, it is
important to know the elementary properties of them well. That these
fields are extremal could be an important ingredient in the
determination of their structure theory, which in turn is an essential
tool in the proof of model theoretic properties.

The notion of "tame valued field" and their model theoretic properties
play a crucial role in the characterization of extremal fields. A valued
field K with separable-algebraic closure K^sep is tame if it is
henselian and the ramification field of the extension K^sep|K coincides
with the algebraic closure. Open problems in the classification of
extremal fields have recently led to new insights about elementary
equivalence of tame fields in the unequal characteristic case. This led
to a follow-up paper. Major suggestions from the referee were worked out
jointly with Sylvy Anscombe and led to stunning insights about the role
of extremal fields as ``atoms'' from which all aleph_1-saturated valued
fields are pieced together.

I will report on joint work with Arno Fehm in which we apply
our previous `existential transfer' results to the problem of
determining which fields admit diophantine nontrivial henselian
valuation rings and ideals. Using our characterization we are able to
re-derive all the results in the literature. Also, I will explain a
connection with Pop's large fields.

How can we explain the patterns of genetic variation in the world around us? The genetic composition of a population can be changed by natural selection, mutation, mating, and other genetic, ecological and evolutionary mechanisms. How do they interact with one another, and what was their relative importance in shaping the patterns we see today? 

Whereas the pioneers of the field could only observe genetic variation indirectly, by looking at traits of individuals in a population, researchers today have direct access to DNA sequences. But making sense of this wealth of data presents a major scientific challenge and mathematical models play a decisive role. This lecture will distil our understanding into workable models and explore the remarkable power of simple mathematical caricatures in interrogating modern genetic data.

To book please email @email

Science is giving us unprecedented insight into the big questions that have challenged humanity. Where did we come from? What is the ultimate destiny of the universe? What are the building blocks of the physical world? What is consciousness?

‘What We Cannot Know’ asks us to rein in this unbridled enthusiasm for the power of science. Are there limits to what we can discover about our physical universe? Are some regions of the future beyond the predictive powers of science and mathematics? Are there ideas so complex that they are beyond the conception of our finite human brains? Can brains even investigate themselves or does the analysis enter an infinite loop from which it is impossible to rescue itself? 

To coincide with the launch of his new book of the same title, Marcus du Sautoy will be answering (or not answering) those questions. He will also be signing copies of the book before and after the lecture.

To book please email @email

Would you like to come see some toys?

'Toys' here have a special sense: objects of daily life which you can find or make in minutes, yet which, if played with imaginatively reveal surprises that keep scientists puzzling for a while. We will see table-top demos of many such toys and visit some of the science that they open up. The common theme is singularity.

Tadashi Tokieda is the Director of Studies in Mathematics at Trinity Hall, Cambridge and the Poincaré Professor in the Department of Mathematics, Stanford.

To book please email @email

To book please email @email