Similarity-Based Pattern Recognition: Challenges and Prospects

Traditional pattern recognition techniques are centered around the notion of "feature". According to this view, the objects to be classified are represented in terms of properties that are intrinsic to the object itself. Hence, a typical pattern recognition system makes its decisions by simply looking at one or more feature vectors provided as input. The strength of this approach is that it can leverage a wide range of mathematical tools ranging from statistics, to geometry, to optimization. However, in many real-world applications a feasible feature-based description of objects might be difficult to obtain or inefficient for learning purposes. In these cases, it is often possible to obtain a measure of the (dis)similarity of the objects to be classified, and in some applications the use of dissimilarities (rather than features) makes the problem more viable. In the last few years, researchers in pattern recognition and machine learning are becoming increasingly aware of the importance of similarity information per se. Indeed, by abandoning the realm of vectorial representations one is confronted with the challenging problem of dealing with (dis)similarities that do not necessarily obey the requirements of a metric. This undermines the very foundations of traditional pattern recognition theories and algorithms, and poses totally new theoretical and computational questions.

The SIMBAD project is a EU FP7 project which aims at undertaking a thorough study of several aspects of purely similarity-based pattern analysis and recognition methods, from the theoretical, computational, and applicative perspective. It aims at covering a wide range of problems and perspectives, including supervised and unsupervised learning, generative and discriminative models, and its interest ranges from purely theoretical problems to real-world practical applications.

Topics of interest for contributed papers include (but are not limited to):

• Foundational issues
• Embedding and embeddability
• Graph spectra and spectral geometry
• Indefinite and structural kernels
• Characterization of non-(geo)metric behavior
• Measures of (geo)metric violations
• Learning and combining similarities
• Multiple-instance learning
• Applications

The workshop aims to explore the spectrum of alternative approaches, methodologies and challenges in the area, rather than detailed techniques. Contributions can be of two kinds:

a) position papers that aim to stimulate discussion of the philosophy of approach underpinning the field,

b) individual technical contributions on a focused topic.