Nonparametric mixtures of factor analyzers

The mixtures of factor analyzers (MFA) model allows data to be modeled as a mixture of Gaussians with a reduced parametrization. We present the formulation of a nonparametric form of the MFA model, the Dirichlet process MFA (DPMFA). The proposed model can be used for density estimation or clustering of high dimensiona data. We utilize the DPMFA for clustering the action potentials of different neurons from extracellular recordings, a problem known as spike sorting. DPMFA model is compared to Dirichlet process mixtures of Gaussians model (DPGMM) which has a higher computational complexity. We show that DPMFA has similar modeling performance in lower dimensions when compared to DPGMM, and is able to work in higher dimensions. ©2009 IEEE.

The dominant role of the solvent-water interface in water droplet templating of polymers

We investigate the formation of microstructured polymer networks known as Breath Figure templated structures created by the presence of water vapour over evaporating polymer solutions. We use a highly controlled experimental approach to examine this dynamic and non-equilibrium process to uniquely compare pure solvent systems with polymer solutions and demonstrate using a combination of optical microscopy, focused ion-beam milling and SEM analysis that the porous polymer microstructure is completely controlled by the interfacial forces that exist between the water droplet and the solvent until a final drying dilation of the imprints. Water droplet contact angles are the same in the presence or absence of polymer and are independent of size for droplets above 5 μm. The polymer acts a spectator that serves to trap water droplets present at the air interface, and to transfer their shape into the polymer film. For the smallest pores, however, there are unexpected variations in the contact angle with pore size that are consistent with a possible contribution from line tension at these smaller dimensions. © The Royal Society of Chemistry.