# Steve Roberts

Steve Roberts is the Director of the Oxford-Man Institute and Professor of Information Engineering at the University of Oxford. He studied physics, completed a PhD in Signal Processing and was appointed to the faculty at Imperial College London, before taking up his post in Oxford in 1999.

He heads the Pattern Analysis and Machine Learning Research Group in the Department of Engineering Science at Oxford.

His main area of research lies in machine learning approaches to data analysis. He has particular interests in the development of machine learning theory for problems in time series analysis and decision theory.

Current research applies Bayesian statistics, graphical models and information theory to diverse problem domains including mathematical biology, finance and sensor fusion.

He has been awarded two medals by the IEE for papers on Bayesian signal analysis.

His current research focuses on statistical models for sequential change-point analysis, forecasting and decision making and decentralised multi-agent co-ordination.

## Related Events

## Working Paper

*Multivariate time series forecasting in incomplete environments*.

*Predicting economic indicators from web text using sentiment composition*.

*Discovering latent association structure via Bayesian one-mode projection of temporal Bipartite graphs*.

*Multi-agent planning with mixed-integer programming and adaptive interaction constraint generation*.

*Predicting economic indicators from web text using sentiment composition*.

*Stochastic processes and feedback-linearisation for online identification and Bayesian adaptive control*.

*Automated machine learning on big data using stochastic algorithm tuning*.

*Communication communities in MOOCs*.

*Variational inference for Gaussian process modulated Poisson processes*.

*Scalable nonparametric Bayesian inference on point processes with Gaussian processes*.

*Soft partitioning in networks via Bayesian non-negative matrix factorization*.

*Bayesian optimisation of Gaussian processes for identifying the deteriorating patient*.

*Identifying sources of discrimination risk in the life cycle of machine intelligence applications under new European Union regs*.

*Improved stochastic trace estimation using mutually unbiased bases*.

*Latent point process allocation*.

*Bayesian Gaussian processes for identifying the deteriorating patient*.

*A direct mapping of Max k-SAT and high order parity checks to a Chimera Graph*.

*Adaptive Bayesian optimisation for online portfolio selection (NIPS)*.

*Pre-earnings announcement drift: Inferring informed trading from the tape*.

*Bayesian inference of log determinants*.

*Practical Bayesian Optimization for Variable Cost Objectives*.

*Distribution of Gaussian process arc lengths. Proceedings of AISTATS*.

*Mosquito detection with neural networks: the buzz of deep learning*.

*Bayesian Heatmaps: probabilistic classification with multiple unreliable information sources*.

*Deep ordinal regression with recurrent neural networks*.

*A novel approach to forecasting financial volatility with Gaussian process envelopes*.

*Entropic trace estimates for log determinants*.

*An information and field theoretic approach to the grand canonical ensemble*.

*p-Markov Gaussian processes for scalable and expressive online Bayesian nonparametric time series forecasting*.

*Generalized spectral kernels*.

*String Gaussian processes*.

*Detecting bird sound in unknown acoustic background using crowdsourced training data*.

*A variational Bayesian state-space approach to online passive-aggressive regression*.

*Predicting dynamic renyi entropy using gaussian processes to estimate financial information flows*.

*Optimal client recommendation for market makers in illiquid financial products*.

## Published Research

*Monthly Notices of the Royal Astronomical Society*. 435 (4). 3639-3653.

*The Astrophysical Journal*. Forthcoming. TBC.

*Animal Behaviour*. 84 (1). 219-223.

*ECML workshop on Scalable Methods in Decision Making*. 23 September 2013.

*Novel Methods for Learning and Optimization of Control Policies and Trajectories for Robotics*. ICRA.

*Advances in Machine Learning for Sensorimotor Control*. NIPS. 2013.

*ISRN Signal Processing*. Article 434832. May 2013.

*Sixth Annual Symposium on Combinatorial Search*. SoCs. 2013.

*International Conference on Machine Learning and Data Mining*. MLDM-13 . To Appear.

*Journal of The Royal Society Interface*. 8 (55). 210-219.

*Monthly Notices of the Royal Astronomical Society*. Forthcoming. Forthcoming.

*Proceedings of uncertainty in artificial intelligence (UAI)*. Forthcoming. Forthcoming.

*Advances in Neural Information Processing Systems (NIPS)*. Forthcoming. Forthcoming.

*Communications of the ACM*. 57 (12). 80-88.

*Journal of Machine Learning Research*. 15. 2337-2397.

*Proceedings of ICICA-2014*. Forthcoming. Forthcoming.

*International Conference on Autonomous Agents and Multiagent Systems*. N/A. N/A.

*Trends in Ecology and Evolution*. 28 (9). 541-551.

*ECML workshop on scalable methods in decision making*. N/A. N/A.

*Novel Methods for Learning and Optimization of Control Policies and Trajectories for Robotics ICRA*. N/A. N/A.

*Advances in Machine Learning for Sensorimotor Control, NIPS*. -. -.

*The Astrophysical Journal*. -. -.

*Monthly Notices of the Royal Astronomical Society*. 428. 2029-2038.

*ISRN Signal Processing*. -. -.

*Sixth Annual Symposium on Combinatorial Search SocS*. -. -.

*International Conference on Machine Learning and Data Mining MLDM-13*. -. -.

*12th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2013)*. -. -.

*Decision Making and Imperfection. Intelligent Systems Reference Library Series*. 474. -.

*Neural Networks*. 24 (7). 726-734.

*Monthly Notices of the Royal Astronomical Society*. 415 (1). 251-256.

*Ad Hoc Networks*. 9 (2). 180-188.

*Physical Review E*. 83 (6). tbc.

*2011 Proceedings of the 14th International Conference on Information Fusion*. 1-8. tbc.

*Super-Resolution Imaging*. tbc. tbc.

*Advances in Ad Hoc Networks*. 9 (2). 180-188.

*Artificial intelligence Review*. tbc. tbc.

*IEEE Signal Processing Letters*. 17 (8). 707-710.

*Journal of the Royal Statistical Society*. 59 (1). 163-173.

*The Computer Journal*. 53 (9). 1430-1446.

*IEEE Signal Processing Letters*. 17 (5). 493-496.

*The Computer Journal*. 53 (9). 1415-1429.

*Pattern Recognition*. 43 (3). 897-905.

*30th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering*. tbc. tbc.

*Proceedings of Autonomous Agents and Multiagent Systems*. tbc. tbc.

*2010 Proceedings of the 13th Conference on Information Fusion*. tbc. 1-9.

*Proceedings of IPSN 2010*. tbc. tbc.

*IEEE Transactions on Aerospace and Electronic Systems*. 46 (1). 207-221.

*The Computer Journal*. 52 . 101-113.

*12 International Conference on Information Fusion*. tbc. 1695-1703.

*Proceedings of the 47th IEEE Conference on Decision and Control*. tbc. 1702-1707.

*Proceedings of the Second AAAI Symposium on Quantum Interaction*. tba. tba.

*Proceedings of the 19th International Conference on Pattern Recognition*. tba. tba.

*Proceedings of the 7th International Conference on Information Processing in Sensor Networks*. tba. 109-120.

*Journal of Artificial Intelligence Research*. 33. 259-283.

*ICA Research Network International Workshop Proceedings*. tba. tba.

*IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems Proceedings*. tba. tba.

*Algorithmic Finance*. 5(1-2). 21-30.

*Monthly Notices of the Royal Astronomical Society*. 467(2). 1661-1677.

*Journal of AI Research*. 57. 661-708.

*Journal of Machine Learning Research*. 17(131). 1-87.

*Monthly Notices of the Royal Astronomical Society*. 462(1). 726-739.

*Monthly Notices of the Royal Astronomical Society*. 456(2). 1374-1393.

*Monthly Notices of the Royal Astronomical Society*. 456(1). L6-L10.

*A Gaussian process framework for modelling stellar activity signals in radial velocity data*.

*Monthly Notices of the Royal Astronomical Society*. 452(2). 1254-1262.