Dr Samer A. Abdallah

Contact Details

Title: Research Assistant
Tel: Internal: [13] 5528
National: 020 7882 5528
International: +44 20 7882 5528
Fax: National: 020 7882 7997
International: +44 20 7882 7997
Email: samer.abdallah@elec.qmul.ac.uk
Room: Eng. 112

Research Group : Centre for Digital Music

Research interests

My research is aimed at the computational modelling of perceptual mechanisms, specifically those involved in audition and music perception. This means constructing artificial systems capable of detecting the structural features of a piece of music, the goal being to produce a meaningful description of the music in terms of relevant structures such as notes, rhythms, harmonies and instrument identities. I am especially interested in the mechanisms whereby this may be accomplished without providing the system with any prior musical knowledge or an external teacher---that is, to replicate the phenomenon of unsupervised perceptual learning.

One possible guiding principle for finding structured patterns in sensory data is that we should seek a description of the input as a combination of statistically independent factors. This has an appealing interpretation as a search for the independent causes of the sensory input. Firstly, the idea of causality has the notion of independence built into it---if two phenomena show some dependence, we are prompted to search either for some causal link between them, or a common cause for them both. It is only when we reach an explanation in terms of independent influences that we can be satistfied that there is no more investigative work to be done. Secondly, the causes of sensory signals are objects in the world, and by seeking causes, we connect with those objects.

Like many researchers working in other areas of cognitive science, I initially approached the problem within a framework of artificial neural networks, but more recently, I have become more interested in the general notion of computation in probabilistic networks, as exemplified by the graphical model formalism. This means that we are less concerned about biologically accurate neural models, and more concerned with the computational principles which might underly them. Fundamental to this is the notion of computation in the face of uncertainty---hence the relevance of probabilistic models, especially Bayesian networks, which are suitable for representing causal structure.. It is interesting to note that graph-structured probabilistic models can support distributed computation and could thus concievably be implemented efficiently within a neural-network-like architecture.

My work so far has involved the use of independent component analysis or ICA for the representation of musical sounds; the use of sparse coding as the basis of an automatic transcription system; and a new approach to event detection based on statistical modelling using ICA and hidden Markov models (HMMs).

Publications

S. A. Abdallah and M. D. Plumbley. Predictive information, Multi-information, and Binding Information. Technical Report C4DM-TR10-10, Centre for Digital Music, Queen Mary University of London, 2010.

S. Abdallah and M. D. Plumbley. Predictive information and Bayesian surprise in exchangeable random processes. Technical Report C4DM-TR10-09, Centre for Digital Music, Queen Mary, University of London. 31 October 2010

S. A. Abdallah and M. D. Plumbley. Information Dynamics: Patterns of expectation and surprise in the perception of music. Connection Science, 21(2), 89-117, 2009. doi: 10.1080/09540090902733756 [Download preprint (930k). This is a preprint of an article whose final and definitive form has been published in Connection Science © 2009 Taylor & Francis; Connection Science is available online at: http://www.informaworld.com/smpp/. OpenURL.]

S. A. Abdallah and M. D. Plumbley. Information dynamics and the perception of temporal structure. In Connectionist Models of Behavior and Cognition II: Proceedings of the 11th Neural Computation and Psychology Workshop, Oxford, UK, 16-18 July 2008, (J. Mayor, N. Ruh and K. Plunkett, eds.), pp179-190. Progress in Neural Processing, Vol. 18, World Scientific Publishing, 2009.

Samer Abdallah. A critique of Dubnov’s ‘information rate’. Technical Report C4DM-TR-08-11, Centre for Digital Music, Queen Mary University of London, 2008.

M. G. Jafari, E. Vincent, S. A. Abdallah, M. D. Plumbley and M. E. Davies. An adaptive stereo basis method for convolutive blind audio source separation. Neurocomputing 71, 2087-2097, 2008.
doi: 10.1016/j.neucom.2007.08.029
[Download Accepted Authors Manuscript (350k) or preprint: pdf (330k)]

J.-B. Thiebaut, S. Abdallah, A. Robertson, N. Bryan-Kinns and M. D. Plumbley. Real time gesture learning and recognition: Towards automatic categorization. In Proceedings of the 8th International Conference on New Interfaces for Musical Expression (NIME 2008), Genova, Italy, 5-7 June 2008, pp 215-217, 2008.

S. A. Abdallah and M. D. Plumbley. Information Dynamics. Technical Report C4DM-TR07-01, Centre for Digital Music, Queen Mary University of London, 18 July 2007.

S A Abdallah and M D Plumbley (2007) Information dynamics and temporal structure in music. Presented at: Language and Music as Cognitive Systems, Cambridge, UK, 11-13 May 2007.

Y. Nishimori, S. Akaho, S. A. Abdallah and M. D. Plumbley. Flag manifolds for subspace ICA problems. In Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2007), Honolulu, Hawaii, USA, 15-20 April 2007. Vol IV, pp IV-1417-1420, 2007.

M. D. Plumbley and S. A. Abdallah. Information theory and sensory perception. In: J. A. Bryant, M. A. Atherton, M. W. Collins (Eds.) Design and Information in Biology: From Molecules to Systems, Chapter 7, pp 205-233. WIT Press, Southampton, UK, 2007. ISBN: 978-1-85312-853-0.

Samer Abdallah, Mark Sandler, Christophe Rhodes, Michael Casey. Using duration models to reduce fragmentation in audio segmentation. Machine Learning, 65, 485 - 515, 2006

S. A. Abdallah and M. D. Plumbley. Geometric Dependency Analysis. Technical Report C4DM-TR-06-05, Centre for Digital Music, Queen Mary University of London, 11 August 2006.
[Download: pdf (1.1M)]

S. A. Abdallah and M. D. Plumbley. Unsupervised analysis of polyphonic music using sparse coding. IEEE Transactions on Neural Networks, 17(1), 179-196, January 2006.
[Supporting material]

M. G. Jafari, S. A. Abdallah, M. D. Plumbley and M. E. Davies. Sparse coding for convolutive blind audio source separation. In Proceedings of the 6th International Conference on Independent Component Analysis and Blind Source Separation (ICA 2006), Charleston, SC, USA, 5-8 March 2006, LNCS 3889, pp 132-139, Springer-Verlag, Berlin, 2006.

M. G. Jafari, E. Vincent, S. A. Abdallah, M. D. Plumbley and M. E. Davies. Blind source separation of convolutive audio using an adaptive stereo basis. To appear in: Proceedings of the ICA Research Network International Workshop (ICArn 2006), 18-19 Sept 2006.

M. G. Jafari, E. Vincent, S. A. Abdallah, M. D. Plumbley and M. E. Davies. An adaptive stereo basis method for convolutive blind audio source separation. Technical Report C4DM-TR-06-04, Centre for Digital Music, Queen Mary University of London, 9 August 2006.

M. D. Plumbley, S. A. Abdallah, T. Blumensath, M. G. Jafari, A. Nesbit, E. Vincent and B. Wang. Musical audio analysis using sparse representations. To appear in Proceedings of 17th COMPSTAT 2006 (17th Symposium of IASC-ERS), Rome, Italy, Aug-Sept 2006.

M. D. Plumbley, S. A. Abdallah, T. Blumensath and M. E. Davies. Sparse Representations of Polyphonic Music. Signal Processing, 86(3), 417-431, March 2006.

E. Vincent, M. G. Jafari, S. A. Abdallah, M. D. Plumbley and M. E. Davies. Blind Audio Source Separation. Technical Report C4DM-TR-05-01. Centre for Digital Music, Queen Mary University of London, 24 November 2005

S. A. Abdallah and M. D. Plumbley. Polyphonic transcription by non-negative sparse coding of power spectra. In Proceedings of the 5th International Conference on Music Information Retrieval (ISMIR 2004), pp 318-325, Barcelona, Spain, October 10-14, 2004.

J. P. Bello, L. Daudet, S. A. Abdallah, C. Duxbury, M. Davies and M. B. Sandler: A Tutorial on Onset Detection in Music Signals. IEEE Transactions on Speech and Audio Processing, 13(5), 1035-1047, Sept. 2005.

S. A. Abdallah and M. D. Plumbley. Application of Geometric Dependency Analysis to the Separation of Convolved Mixtures. In 5th International Conference on Independent Component Analysis and Blind Signal Separation (ICA 2004), pp 318-325, Granada, Spain, September 22-24, 2004.

Samer A. Abdallah and Mark D. Plumbley: Unsupervised Onset Detection: a Probabilistic Approach using ICA and a Hidden Markov Classifier. In Cambridge Music Processing Colloquium, Cambridge, UK, 2003.

Samer A. Abdallah and Mark D. Plumbley: Probability as metadata: Event detection in music using ICA as a conditional density model. In Proc. 4th Intl. Symp. on Independent Component Analysis and Signal Separation (ICA2003), pp 233-238, Nara, Japan, 2003.

Samer A. Abdallah and Mark D. Plumbley: Geometric ICA Using Nonlinear Correlation and MDS. In Proc. 4rd Intl. Symp. on Independent Component Analysis and Signal Separation (ICA2003), pp 161-166, Nara, Japan, 2003.

Samer A. Abdallah and Mark D. Plumbley: An Independent Component Analysis Approach to Automatic Music Transcription. In Proceedings of the 114th Convention of the Audio Engineering Society, Amsterdam, March 2003.

Samer A. Abdallah: Towards Music Perception by Redundancy Reduction and Unsupervised Learning in Probabilistic Models. PhD thesis, Department of Electronic Engineering, King's College London, 2002.

Mark D. Plumbley, Samer A. Abdallah, Juan P. Bello, Mike E. Davies, Giuliano Monti, and Mark B. Sandler: Automatic Music Transcription and Audio Source Separation. Cybernetics and Systems, 33(6):603-627, 2002.
[Download preprint: article (pdf: 52k) and figures (pdf: 315k). NB Fig 8 may render very slowly.]

Samer A. Abdallah and Mark D. Plumbley: If the independent components of natural scenes are edges, what are the independent components of natural sounds? In Proc. 3rd Intl. Conf. on Independent Component Analysis and Signal Separation, ICA2001, pages 534-539, San Diego, 2001.

Mark D. Plumbley, Samer A. Abdallah, Juan P. Bello, Mike E. Davies, J. Klingseisen, Giuliano Monti, and Mark B. Sandler: ICA and Related Models Applied to Audio Analysis and Separation. In Proceedings of the Fourth International ICSC Symposium on Soft Computing and Intelligent Systems for Industry, Paisley, Scotland, 2001.

Samer A. Abdallah: Music Perception Through Unsupervised Learning. Technical report, Department of Electronic Engineering, King's College London, 1999.

Samer A. Abdallah and Mark D. Plumbley: Unsupervised Learning for Music Perception. In Proc. Cambridge Music Processing Colloquium, Cambridge, UK, 1999.