We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

IEEE Transactions on Pattern Analysis and Machine Intelligence

Deciding What's News: A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time

We combine supervised learning with unsupervised learning in deep neural networks. The proposed model is trained to simultaneously minimize the sum of supervised and unsupervised cost functions by backpropagation, avoiding the need for layer-wise pre-training. Our work builds on top of the Ladder network proposed by Valpola [1] which we extend by combining the model with supervision. We show that the resulting model reaches state-of-the-art performance in semi-supervised MNIST and CIFAR-10 classification in addition to permutation-invariant MNIST classification with all labels.

https://proceedings.neurips.cc/paper/2015/file/378a063b8fdb1db941e34f4bde584c7d-Paper.pdf

Semi-supervised learning with Ladder networks

L'article, publie dans Science, sur une des premieres utilisations analytiques de Google Books, fondee sur les n-grammes (Google Ngrams) We constructed a corpus of digitized texts containing about 4% of all books ever printed. Analysis of this corpus enables us to investigate cultural trends quantitatively. We survey the vast terrain of "culturomics", focusing on linguistic and cultural phenomena that were reflected in the English language between 1800 and 2000. We show how this approach can ...

Quantitative Analysis of Culture Using Millions of Digitized Books

We tackle the problem of online reward maximisation over a large finite set of actions described by their contexts. We focus on the case when the number of actions is too big to sample all of them even once. However we assume that we have access to the similarities between actions' contexts and that the expected reward is an arbitrary linear function of the contexts' images in the related reproducing kernel Hilbert space (RKHS). We propose KernelUCB, a kernelised UCB algorithm, and give a cumulative regret bound through a frequentist analysis. For contextual bandits, the related algorithm GP-UCB turns out to be a special case of our algorithm, and our finite-time analysis improves the regret bound of GP-UCB for the agnostic case, both in the terms of the kernel-dependent quantity and the RKHS norm of the reward function. Moreover, for the linear kernel, our regret bound matches the lower bound for contextual linear bandits.

Finite-Time Analysis of Kernelised Contextual Bandits

News content analysis is usually preceded by a labour-intensive coding phase, where experts extract key information from news items. The cost of this phase imposes limitations on the sample sizes that can be processed, and therefore to the kind of questions that can be addressed. In this paper we describe an approach that incorporates text-analysis technologies for the automation of some of these tasks, enabling us to analyse data sets that are many orders of magnitude larger than those normally used. The patterns detected by our method include: (1) similarities in writing style among several outlets, which reflect reader demographics; (2) gender imbalance in media content and its relation with topic; (3) the relationship between topic and popularity of articles.

https://www.tandfonline.com/doi/pdf/10.1080/21670811.2012.714928

Research methods in the age of digital journalism

Machine Learning and Knowledge Discovery in Databases (ECML PKDD)

. In this study we present a new cyclone identification and tracking algorithm, cycloTRACK. The algorithm describes an iterative process. At each time step it identifies all potential cyclone centers, defined as relative vorticity maxima embedded in smoothed enclosed contours of at least 3 × 10−5 s−1 at the atmospheric level of 850 hPa. Next, the algorithm finds all the potential cyclone paths by linking the cyclone centers at consecutive time steps and selects the most probable track based on the minimization of a cost function. The cost function is based on the average differences of relative vorticity between consecutive track points, weighted by their distance. Last, for each cyclone, the algorithm identifies "an effective area" for which different physical diagnostics are measured, such as the minimum sea level pressure and the maximum wind speed. The algorithm was applied to the ERA-Interim reanalyses for tracking the Northern Hemisphere extratropical cyclones of winters from 1989 until 2009, and we assessed its sensitivity for the several free parameters used to perform the tracking.

/pdf/cyclotrack-v1-0-tracking-winter-extratropical-cyclones-based-1u4qofp6af.pdf

CycloTRACK (v1.0) – tracking winter extratropical cyclones based on relative vorticity: sensitivity to data filtering and other relevant parameters

/pdf/finite-time-analysis-of-kernelised-contextual-bandits-3gptr0m114.pdf

Ilias Flaounas

Papers

Finite-Time Analysis of Kernelised Contextual Bandits

Research methods in the age of digital journalism

Machine Learning and Knowledge Discovery in Databases (ECML PKDD)

CycloTRACK (v1.0) – tracking winter extratropical cyclones based on relative vorticity: sensitivity to data filtering and other relevant parameters

Finite-time analysis of kernelised contextual bandits