E. V. Bonilla, Felix V. Agakov, C. K. I. Williams.
Kernel Multi-task Learning using Task-specific Features.
To appear in AISTATS 2007, The Society for Artificial Intelligence and Statistics, 8p. [ pdf]

J. Cavazos, C. Dubach, F. V. Agakov, E. V. Bonilla, M. F. P. O'Boyle, G. Fursin, O. Temam.
Automatic Performance Model Construction for the Fast Software Exploration of New Hardware Designs.
In Proc. Int. Conf. on Compilers, Architecture, and Synthesis for Embedded Systems (CASES), ACM, 2006, 11p. [ pdf]

E. V. Bonilla, C. K. I. Williams, F. V. Agakov, J. Cavazos, J. Thomson, M. F. P. O'Boyle.
Predictive Search Distributions.
In Proc. ICML, OmniPress, 2006, 8p. [pdf]

F. V. Agakov, E. V. Bonilla, J. Cavazos, B. Franke, G. Fusin, M. F. P. O'Boyle, J. Thomson, M. Toussaint, C. K. I. Williams.
Using Machine Learning to Focus Iterative Optimization.
In the 4th Annual International Symposium on Code Generation and Optimization (CGO), IEEE Comp. Soc., 2006, 11p. [pdf]

F. V. Agakov and D. Barber. Auxiliary Variational Information Maximization for Dimensionality Reduction.
In C. Saunders, M. Grobelnik, S. Gunn, J. Shawe-Taylor (Eds.),
Revised selected papers of Subspace, Latent Structure, and Feature Selection Workshop (SLSFS), LNCS 3940, Springer, 2006, 12p. [ps.gz, pdf]

F.V. Agakov and D. Barber. Kernelized Infomax Clustering.
In Neural Information Processing Systems 18, MIT Press, 2005, 8p. [ps.gz, pdf]

F. V. Agakov and D. Barber. Variational Information Maximization for Neural Coding.
In International Conference on Neural Information Processing, Springer, 2004, 6p. [pdf]

F. V. Agakov and D. Barber. An Auxiliary Variational Method.
In International Conference on Neural Information Processing, Springer, 2004, 6p. [pdf]

D. Barber and F. V. Agakov. The IM Algorithm: A variational approach to Information Maximization.
In Neural Information Processing Systems 16, MIT Press, 2003, 8p. [ps.gz, pdf]

M. Welling, F. V. Agakov and C. K. I. Williams. Extreme Components Analysis.
In Neural Information Processing Systems 16, MIT Press, 2003, 8p. [ps.gz, pdf]

F. V. Agakov and D. Barber. Approximate Learning in Temporal Hidden Hopfield Models.
In International Conference on Artificial Neural Networks, Springer-Verlag, 2003, 8p. [ps.gz, pdf].

C. K. I. Williams and F. V. Agakov. Products of Gaussians and Probabilistic Minor Component Analysis.
Neural Computation, Vol. 14, No. 5, MIT Press, 2002. [ps.gz]

C. K. I. Williams, F. V. Agakov and S. N. Felderhof. Products of Gaussians.
In Neural Information Processing Systems 14, MIT Press, 2001, 8p. [ps.gz]

Research reports:

F. V. Agakov and D. Barber. Variational Information Maximization in Gaussian Channels.
UoE, IANC. Technical Report, EDI-INF-RR-0206, Apr 2004, 12p. [ps.gz, pdf]

F. V. Agakov and D. Barber. An Auxiliary Variational Method.
UoE, IANC. Technical Report, EDI-INF-RR-0205, Apr 2004, 10p. [ps.gz, pdf]

F. V. Agakov and D. Barber. Temporal Hidden Hopfield Models.
UoE, IANC. Technical Report, EDI-INF-RR-0156, Nov 2002, 18p. [ps.gz, pdf]
(Here are the slides for the related talk I gave at the Probabilistic Brain symposium in Cambridge, UK '03).

D. Barber and F. V. Agakov. Correlated sequence learning in a network
of spiking neurons using maximum likelihood.
UoE, IANC. Technical Report, EDI-INF-RR-0149, Apr 2002, 13p. [ps.gz]

C. K. I. Williams and F. V. Agakov. An Analysis of Contrastive Divergence
Learning in Gaussian Boltzmann Machines.
UoE, IANC. Technical Report, EDI-INF-RR-0120, May 2002, 14p. [ps.gz]

C. K. I. Williams and F. V. Agakov. Products of Gaussians and Probabilistic Minor Component Analysis.
UoE, Institute of Adaptive and Neural Computation. Technical Report, EDI-INF-RR-0043, July 2001, 11p.
[ps.gz]

---

Refereed workshops:

F. V. Agakov and D. Barber. Information-Theoretic Clustering in Nonlinear Encoder Models.
In PASCAL: Statistics and Optimization of Clustering Workshop, 2005. [pdf]

F. V. Agakov and D. Barber. Auxiliary Variational Information Maximization for Dimensionality Reduction.
In PASCAL: Subspace, Latent Structure and Feature Selection Techniques: Statistical and Optimisation Perspectives Workshop, 2005. [ps.gz, pdf]

---

Theses:

F. V. Agakov. Variational Information Maximization in Stochastic Environments.
PhD Thesis, School of Informatics, The University of Edinburgh, 2005, 205p. [pdf]

F. V. Agakov. Investigations of Gaussian Products-of-Experts Models.
Master's Thesis, Division of Informatics, The University of Edinburgh, 2000, 169p. [ps.gz]
(See also introductory [ps.gz] and even more introductory [ps.gz] slides).
Here are updates of supporting calculations for the variance of the weight update in contrastive divergence learning in a simple 1-factor 1-D factor analysis model [ps.gz].

Here is the list of older publications.

---

Back to the main page.

---

This page is maintained by Felix Agakov

Page last updated: Feb. 11, 2007