Moving the goal posts: a reply to Dawson and Piercey

Berkeley [Minds Machines 10 (2000) 1] described a methodology that showed the subsymbolic nature of an artificial neural network system that had been trained on a logic problem, originally described by Bechtel and Abrahamsen [Connectionism and the mind. Blackwells, Cambridge, MA, 1991]. It was also claimed in the conclusion of this paper that the evidence was suggestive that the network might, in fact, count as a symbolic system. Dawson and Piercey [Minds Machines 11 (2001) 197] took issue with this latter claim. They described some lesioning studies that they argued showed that Berkeley’s (2000) conclusions were premature. In this paper, these lesioning studies are replicated and it is shown that the effects that Dawson and Piercey rely upon for their argument are merely an artifact of a threshold function they chose to employ. When a threshold function much closer to that deployed in the original studies is used, the significant effects disappear.     

Touch‐screen panel integrated into 12.1‐in. a‐Si:H TFT‐LCD

Abstract— A touch‐screen‐panel (TSP) embedded 12.1‐in. LCD employing a standard existing a‐Si:H TFT‐LCD process has been successfully developed. Compared with conventional external touch‐screen panels, which use additional components to detect touch events, the new internal TSP exhibits a clearer image and improved touch feeling, as well as increased sensing speed using discrete sensing lines to enable higher‐speed sensing functions including handwriting. The new internal digital switching TSP can be fabricated with low cost because it does not require any additional process steps compared to a standard a‐Si:H TFT‐LCD.     

LED‐backlight feedback control system with integrated amorphous‐silicon color sensor on an LCD panel

Abstract— Thin‐film‐transistor liquid‐crystal displays (TFT‐LCDs) have the largest market share of all digital flat‐panel displays. An LCD backlighting system employing a three‐color red‐green‐blue light‐emitting diode (RGB‐LED) array is very attractive, considering its wide color gamut, tunable white point, high dimming ratio, long lifetime, and environmental compatibility. But the high‐intensity LED has problems with thermal stability and degradation of brightness over time. Color and white luminance levels are not stable over a wide range of temperature due to inherent long‐term aging characteristics. In order to minimize color point and brightness differences over time, optical feedback control is the key technology for any LED‐backlight system. In this paper, the feasibility of an optical color‐sensing feedback system for an LED backlight by integrating the amorphous‐silicon (a‐Si) color sensor onto the LCD panel will be presented. To minimize the photoconductivity degradation of a‐Si, a new laser exposure treatment has been applied. The integrated color‐sensor optical‐feedback‐controlled LED‐backlight system minimized the color variation to less than 0.008 Δu'v' (CIE1976) compared to 0.025 for an open‐loop system over the temperature range of 42–76°C.     

Peter Novak,Mental Symbols: A Defence of the Classical Theory of Mind. Studies in Cognitive Systems 19, Dordrecht,Netherlands: Kluwer Academic Publishers, 1997, xxii + 266 pp., $114.00, ISBN 0-7923-4370-0.

Minds and Machines -     

Novel 120‐Hz TFT‐LCD motion‐blur‐reduction technology with integrated motion‐compensated frame‐interpolation timing controller

Abstract— Samsung has developed a high‐resolution full‐HD (1920 × 1080) 120‐Hz LCD‐TV panel using a novel pixel structure and a motion‐compensated frame‐interpolation (McFi) single‐chip solution. Our latest work includes launch of a 70‐in. full‐HD panel, the world's largest LCD TV in mass production, with a 120‐Hz frame rate. A serious problem involving the charging time margin has been completely overcome through the use of a new alternative 1G‐2D pixel structure and a new driving scheme. Compared with conventional dot‐inversion driving, our new dot‐inversion method, which is a spatial averaging technique, can save power because the column drivers are operated using vertical inversion driving. In addition, McFi, which merges individual ME/MC and timing‐controller (TCON) ICs and memories, has been developed and applied in a mass‐production product for the first time ever. The McFi solution provides 120‐Hz driving with the lowest possible system cost. Motion‐picture response time (MPRT) has been reduced from 1 5 to 8 msec. Moreover, for the case of 24‐Hz film source mode, motion judder has been completely eliminated. As a result, a lineup consisting of 40‐, 46‐, 52‐, 70‐, and 82‐in. LCD‐TV panels with high quality and manufacturability has been made possible.     

Conducting banding analysis with trained networks of sigmoid units

Berkeley et al. (1995, Connection Science, 7: 167–186) introduced a novel technique for analysing the hidden units of connectionist networks that had been trained using the backpropagation learning procedure. The literature concerning banding analysis is equivocal with respect to the kinds of processing units this technique can be used on. In this paper, it will be shown that, contrary to the claims in some published sources, banding analysis can be conducted on networks that use standard processing units that have a sigmoid activation function. The analytic process is then illustrated and the potential benefits of this kind of technique are discussed.     

An old-fashioned connectionist approach to a Cajun chord change problem

In this paper, the problem of changing chords when playing Cajun music is introduced. A number of connectionist network simulations are then described, in which the networks attempted to learn to predict chord changes correctly in a particular Cajun song, ‘Bayou Pompon’. In the various sets of simulations, the amount of information provided to the network was varied. While the network had difficulty in solving the problem with six one-eighths of a bar of melody information, performance radically improved when the network was provided with seven one-eighths of a bar of melody information. A post-training analysis of a trained network revealed a ‘rule’ for solving the problem. In addition to providing useful insight for scholars interested in traditional Cajun music, the results described here also illustrate how a traditional connectionist network, trained with the familiar backpropagation learning algorithm, can be used to generate a theory of the task.     

Multi-Material ALE with AMR for Modeling Hot Plasmas and Cold Fragmenting Materials

We have developed a new 3D multi-physics multi-material code,ALE-AMR,which combines Arbitrary Lagrangian Eulerian(ALE) hydrodynamics with Adaptive Mesh Refinement(AMR) to connect the continuum to the microstructural regimes.The code is unique in its ability to model hot radiating plasmas and cold fragmenting solids.New numerical techniques were developed for many of the physics packages to work efficiently on a dynamically moving and adapting mesh.We use interface reconstruction based on volume fractions of the material components within mixed zones and reconstruct interfaces as needed.This interface reconstruction model is also used for void coalescence and fragmentation.A flexible strength/failure framework allows for pluggable material models,which may require material history arrays to determine the level of accumulated damage or the evolving yield stress in J2 plasticity models.For some applications laser rays are propagating through a virtual composite mesh consisting of the finest resolution representation of the modeled space.A new 2nd order accurate diffusion solver has been implemented for the thermal conduction and radiation transport packages.One application area is the modeling of laser/target effects including debris/shrapnel generation.Other application areas include warm dense matter,EUV lithography,and material wall interactions for fusion devices.     

Thermally adaptive response‐time compensation for LCDs

Abstract— This paper presents thermally adaptive driving (TAD) technology for response‐time compensation (RTC) of an LCD with an integrated sensor. The TAD system is comprised of an analog sensor, an analog sensor signal conditioning, and a digital feedback algorithm. The integrated thermal sensor provides accurate temperature measurement of the liquid‐crystal layer. The TAD controller has an eight‐step look‐up‐table (LUT) and compensates response time based on the panel temperature. The TAD system reduces response time by nearly 34% over the temperature range 0–60°C. This paper also presents a thermal sensor which has been integrated onto an LCD. The sensor uses metal (Mo/Al) film as a temperature detection layer, and its fabrication requires no manufacturing process changes. The sensor shows very good linearity, sensitivity, and reliability.