
Tighter risk certificates for neural networks
This paper presents empirical studies regarding training probabilistic n...
read it

PACBayes Analysis Beyond the Usual Bounds
We focus on a stochastic learning model where the learner observes a fin...
read it

PACBayes unleashed: generalisation bounds with unbounded losses
We present new PACBayesian generalisation bounds for learning problems ...
read it

Predicting Engagement in Video Lectures
The explosion of Open Educational Resources (OERs) in the recent years c...
read it

Correlated Feature Selection with Extended Exclusive Group Lasso
In many high dimensional classification or regression problems set in a ...
read it

Evolution of a Complex PredatorPrey Ecosystem on Largescale MultiAgent Deep Reinforcement Learning
Simulation of population dynamics is a central research theme in computa...
read it

Towards an Integrative Educational Recommender for Lifelong Learners
One of the most ambitious use cases of computerassisted learning is to ...
read it

TrueLearn: A Family of Bayesian Algorithms to Match Lifelong Learners to Open Educational Resources
The recent advances in computerassisted learning systems and the availa...
read it

Constructing Artificial Data for Finetuning for LowResource Biomedical Text Tagging with Applications in PICO Annotation
Biomedical text tagging systems are plagued by the dearth of labeled tra...
read it

DataDriven Malaria Prevalence Prediction in Large DenselyPopulated Urban Holoendemic subSaharan West Africa: Harnessing Machine Learning Approaches and 22years of Prospecti
Plasmodium falciparum malaria still poses one of the greatest threats to...
read it

Deep Learning Enhanced Extended DepthofField for Thick BloodFilm Malaria HighThroughput Microscopy
Fast accurate diagnosis of malaria is still a global health challenge fo...
read it

Perturbed Model Validation: A New Framework to Validate Model Relevance
This paper introduces PMV (Perturbed Model Validation), a new technique ...
read it

Interlacing Personal and Reference Genomes for Machine Learning DiseaseVariant Detection
DNA sequencing to identify genetic variants is becoming increasingly val...
read it

Structured MultiLabel Biomedical Text Tagging via Attentive Neural Tree Decoding
We propose a model for tagging unstructured texts with an arbitrary numb...
read it

Faster Convergence & Generalization in DNNs
Deep neural networks have gained tremendous popularity in last few years...
read it

Spatiotemporal Prediction of Ambulance Demand using Gaussian Process Regression
Accurately predicting when and where ambulance callouts occur can reduc...
read it

PACBayes bounds for stable algorithms with instancedependent priors
PACBayes bounds have been proposed to get risk estimates based on a tra...
read it

Adaptive Mechanism Design: Learning to Promote Cooperation
In the future, artificial learning agents are likely to become increasin...
read it

Empirical Risk Minimization under Fairness Constraints
We address the problem of algorithmic fairness: ensuring that sensitive ...
read it

Improving Active Learning in Systematic Reviews
Systematic reviews are essential to summarizing the results of different...
read it

A Tutorial on Canonical Correlation Methods
Canonical correlation analysis is a family of multivariate statistical m...
read it

Improved Particle Filters for Vehicle Localisation
The ability to track a moving vehicle is of crucial importance in numero...
read it

Localized Lasso for HighDimensional Regression
We introduce the localized Lasso, which is suited for learning models th...
read it

Learning Shared Representations in Multitask Reinforcement Learning
We investigate a paradigm in multitask reinforcement learning (MTRL) i...
read it

PinView: Implicit Feedback in ContentBased Image Retrieval
This paper describes PinView, a contentbased image retrieval system tha...
read it

PACBayes Analysis of Multiview Learning
This paper presents eight PACBayes bounds to analyze the generalization...
read it

Retrieval of Experiments by Efficient Estimation of Marginal Likelihood
We study the task of retrieving relevant experiments given a query exper...
read it

Challenges in Representation Learning: A report on three machine learning contests
The ICML 2013 Workshop on Challenges in Representation Learning focused ...
read it

MahNMF: Manhattan Nonnegative Matrix Factorization
Nonnegative matrix factorization (NMF) approximates a nonnegative matr...
read it

PACBayesian Inequalities for Martingales
We present a set of highprobability inequalities that control the conce...
read it

PACBayesian Analysis of the ExplorationExploitation Tradeoff
We develop a coherent framework for integrative simultaneous analysis of...
read it

PACBayesian Analysis of Martingales and Multiarmed Bandits
We present two alternative ways to apply PACBayesian analysis to sequen...
read it

Gaussian Process Bandits for Tree Search: Theory and Application to Planning in Discounted MDPs
We motivate and analyse a new Tree Search algorithm, GPTS, based on rece...
read it

A Nonconformity Approach to Model Selection for SVMs
We investigate the issue of model selection and the use of the nonconfor...
read it
John ShaweTaylor
is this you? claim profile
John Stewart ShaweTaylor is the Director of the University College Center for Computational Statistics and Machine Learning. His main field of research is the theory of statistical learning. He has contributed to a variety of areas from graphic theory, cryptography, statistical theory of learning and its applications. His main contributions were, however, in developing the analysis and subsequent algorithm definition of principle algorithms based on the theory of statistical learning. This work has contributed to a major rebirth in machine learning by introducing kernel methods and supporting vector machines, including mapping these approaches to new domains, including computer vision, document graduation and brain scan analysis. He has worked more recently on interactive learning and strengthening learning. He has also played a key role in bringing together a number of important European Networks of Excellence. The scientific coordination of these programs has influenced a generation of researchers and promoted the widespread use of machine learning in science and industry. More than 300 papers with more than 42,000 citations have been published. Two coauthored books with Nello Cristianini have become standard monographs for studying kernel processes and vector supporting machines, attracting 21,000 quotations. He’s head of the department of computer science at University College London, where he has monitored a significant expansion and has been seen emerging from the United Kingdom Research Evaluation Framework as the top computer science department in 2014.