Scaling POMDPs for dialog management with composite summary point-based value iteration (CSPBVI)

Although partially observable Markov decision processes (POMDPs) have shown great promise as a framework for dialog management in spoken dialog systems, important scalability issues remain. This paper tackles the problem of scaling slot-filling POMDP-based dialog managers to many slots with a novel technique called composite point-based value iteration (CSPBVI). CSPBVI creates a "local" POMDP policy for each slot; at runtime, each slot nominates an action and a heuristic chooses which action to take. Experiments in dialog simulation show that CSPBVI successfully scales POMDP-based dialog managers without compromising performance gains over baseline techniques and preserving robustness to errors in user model estimation. Copyright © 2006, American Association for Artificial Intelligence (www.aaai.org). All rights reserved.

Effect of gate dielectric scaling in nanometer scale vertical thin film transistors

A short channel vertical thin film transistor (VTFT) with 30 nm SiN x gate dielectric is reported for low voltage, high-resolution active matrix applications. The device demonstrates an ON/OFF current ratio as high as 10 9, leakage current in the fA range, and a sub-threshold slope steeper than 0.23 V/dec exhibiting a marked improvement with scaling of the gate dielectric thickness. © 2011 American Institute of Physics.

Scaling the Indian Buffet Process via Submodular Maximization

Inference for latent feature models is inherently difficult as the inference space grows exponentially with the size of the input data and number of latent features. In this work, we use Kurihara & Welling (2008)'s maximization-expectation framework to perform approximate MAP inference for linear-Gaussian latent feature models with an Indian Buffet Process (IBP) prior. This formulation yields a submodular function of the features that corresponds to a lower bound on the model evidence. By adding a constant to this function, we obtain a nonnegative submodular function that can be maximized via a greedy algorithm that obtains at least a one-third approximation to the optimal solution. Our inference method scales linearly with the size of the input data, and we show the efficacy of our method on the largest datasets currently analyzed using an IBP model.

Scaling of transitionally-rough flow structures in DNSes of channels with riblets up to Reτ=550

The linear, drag-reducing effect of vanishingly small riblets breaks down once their size is in the transitionally-rough regime. We have previously reported that this breakdown is caused by the additional Reynolds stresses produced by the appearance of elongated spanwise rollers just above the riblet surface. These rollers are related with the Kelvin--Helmholtz instability of free shear layers, and to similar structures appearing over other rough and porous surfaces. However, because of the limited Reτ=180 in our previous DNSes, it could not be determined whether those structures scaled in inner or outer units. Furthermore, it is questionable if results in the transitionally-rough regime at Reτ=180 can be extrapolated to configurations of practical interest. At such small Reynolds numbers, roughness of transitional size can perturb a large portion of the boundary layer, which is not the case in most industrial and atmospheric applications. To clarify these issues we have conducted a set of DNSes at Reτ=550. Our results indicate that the spanwise rollers scale in wall units, and support the validity of the extrapolation to configurations of practical interest.

Growth kinetics and uniform scaling-up of graphene synthesis

Chemical vapor deposition on copper is the most widely used method to synthesize graphene at large scale. However, the clear understanding of the fundamental mechanisms that govern this synthesis is lacking. Using a vertical-flow, cold-wall reactor with short gas residence time we observe the early growths to study the kinetics of chemical vapor deposition of graphene on copper foils and demonstrate uniform synthesis at wafer scale. Our results indicate that the growth is limited by the catalytic dissociative dehydrogenation on the surface and copper sublimation hinders the graphene growth. We report an activation energy of 3.1 eV for ethylene-based graphene synthesis. © The Electrochemical Society.