The Cartesian Test for Automatism1

In Part V of his Discourse on the Method, Descartes introduces a test for distinguishing people from machines that is similar to the one proposed much later by Alan Turing. The Cartesian test combines two distinct elements that Keith Gunderson has labeled the language test and the action test. Though traditional interpretation holds that the action test attempts to determine whether an agent is acting upon principles, I argue that the action test is best understood as a test of common sense. I also maintain that this interpretation yields a stronger test than Turing's, and that contemporary artificial intelligence should consider using it as a guide for future research.     

Structure-based design of inhibitors of purine nucleoside phosphorylase. 3. 9-Arylmethyl derivatives of 9-deazaguanine substituted on the methylene group

X-ray crystallography and computer-assisted molecular modeling (CAMM) studies aided in the design of a potent series of mammalian purine nucleoside phosphorylase (PNP) inhibitors. Enhanced potency was achieved by designing substituted 9-(arylmethyl)-9-deazaguanine analogs that interact favorably with all three of the binding subsites of the PNP active site, namely the purine binding site, the hydrophobic pocket, and the phosphate binding site. The most potent PNP inhibitor prepared during our investigation, (S)-9-[1-(3-chlorophenyl)-2-carboxyethyl]-9-deazaguanine (18b), was shown to have an IC50 of 6 nM, whereas the corresponding (R)-isomer was 30-fold less potent.     

Suppression of substance P biosynthesis in sensory neurons of dorsal root ganglion by prodrug esters of potent peptidylglycine alpha-amidating monooxygenase inhibitors

Substance P as well as many other neuropeptides are synthesized as glycine-extended precursors and converted to the biologically active C-terminal amides by posttranslational modification. The final step of posttranslational processing is catalyzed by peptidylglycine alpha-amidating monooxygenase (PAM). In a previous study, N-substituted homocysteine analogs were found to be potent inhibitors of PAM partially purified from conditioned medium of cultured rat medullary thyroid carcinoma CA-77 cells. These compounds, however, were only modest inhibitors of substance P production in cultured dorsal root ganglion cells, possibly because of poor cell penetration. Several ester derivatives of hydrocinnamoyl-phenylalanyl-homocysteine, one of the most potent PAM inhibitors, were prepared to increase the intracellular accessibility of these compounds. Hydrocinnamoyl-phenylalanyl-(S-benzoyl-homocysteine) benzyl ester was identified as the most potent compound, inhibiting substance P biosynthesis in dorsal root ganglion cells with an IC50 of 2 microM. Inhibition of PAM resulted in a concomitant increase in the glycine-extended substance p (substance P-Gly) precursor peptide. In the presence of 3 microM benzyl ester derivative, the intracellular substance P-Gly level was 2.4-fold higher while the substance P level was 2.1-fold lower than the corresponding peptides in control cells. These results suggest that PAM inhibition represents an effective method for suppression of substance P biosynthesis and, therefore, may have therapeutic utility in conditions associated with elevated substance P levels. Furthermore, PAM inhibition may also prove useful in decreasing other amidated peptides.     

Purine nucleoside phosphorylase inhibitors: biochemical and pharmacological studies with 9-benzyl-9-deazaguanine and related compounds

Certain derivatives of 9-deazaguanine that contain arylmethyl, heteroarylmethyl or cycloalkylmethyl groups at the 9-position are potent inhibitors of purine nucleoside phosphorylase (PNP, E.C. 2.4.2.1). To determine whether these agents can produce metabolically significant inhibition of PNP in cells and in animals, the authors performed pharmacological studies with a representative member of the series, 9-benzyl-9-deazaguanine (BzDAG). BzDAG was a potent inhibitor of PNP from calf spleen (Ki = 12 nM). It was also an effective inhibitor of PNP in cells and in animals as shown by the findings that it 1) inhibited the conversion of inosine to nucleotides in L1210 cells in culture at concentrations that had little effect on the utilization of hypoxanthine; 2) potentiated the toxicity of deoxyguanosine to CCRF-CEM cells in culture; 3) increased the pools of deoxy GTP in CCRF-CEM, Molt-3 and Molt-4 cells that had been treated with deoxyguanosine; 4) prevented the toxicity of 6-thioguanosine to HEp-2 cells in culture; 5) increased the plasma levels of endogenous inosine in rats; and 6) increased the plasma levels of 2',3'-dideoxyinosine in rats that had received BzDAG and dideoxyinosine in combination. Pharmacokinetic analysis of BzDAG in the rat showed it to be 48% orally bioavailable (at a dose of 5 mg/kg). About 95% of BzDAG was protein bound. After i.v. administration of BzDAG (5 mg/kg), more than 50% of the erythrocyte PNP was inhibited for 40 min. These results indicate that the 9-substituted-9-deazaguanines are potent orally active PNP inhibitors and are therefore of potential clinical interest as immunosuppressive and anti-inflammatory agents.     

Distributed computation of the knn graph for large high-dimensional point sets

High-dimensional problems arising from robot motion planning, biology, data mining, and geographic information systems often require the computation of k nearest neighbor (knn) graphs. The knn graph of a data set is obtained by connecting each point to its k closest points. As the research in the above-mentioned fields progressively addresses problems of unprecedented complexity, the demand for computing knn graphs based on arbitrary distance metrics and large high-dimensional data sets increases, exceeding resources available to a single machine. In this work we efficiently distribute the computation of knn graphs for clusters of processors with message passing. Extensions to our distributed framework include the computation of graphs based on other proximity queries, such as approximate knn or range queries. Our experiments show nearly linear speedup with over 100 processors and indicate that similar speedup can be obtained with several hundred processors.     

Hypoglycaemic effect of AICAriboside in mice

We have previously demonstrated that in isolated hepatocytes from fasted rats, AICAriboside (5-amino 4-imidazolecarboxamide riboside), after its conversion into AICAribotide (AICAR or ZMP), exerts a dose-dependent inhibition on fructose-1,6-bisphosphatase and hence on gluconeogenesis. To assess the effect of AICAriboside in vivo, we measured plasma glucose and liver metabolites after intraperitoneal administration of AICAriboside in mice. In fasted animals, in which gluconeogenesis is activated, AICAriboside (250 mg/kg body weight) induced a 50% decrease of plasma glucose within 15 min, which lasted about 3 h. In fed mice, glucose decreased by 8% at 30 min, and normalized at 1 h. Under both conditions, ZMP accumulated to approximately 2 mumol/g of liver at 1 h. It decreased progressively thereafter, although much more slowly in the fasted state. Inhibition of fructose-1,6-bisphosphatase was evidenced by time-wise linear accumulations of fructose-1,6-bisphosphate, from 0.006 to 3.9 mumol/g of liver at 3 h in fasted mice, and from 0.010 to 0.114 mumol/g of liver at 1 h in fed animals. AICAriboside did not significantly influence plasma insulin or glucose utilization by muscle. We conclude that in vivo as in isolated hepatocytes, AICAriboside, owing to its conversion into ZMP, inhibits fructose-1,6-bisphosphatase and consequently gluconeogenesis.     

Domain and requirements for a wearable-based doping control system

In Germany, over 7000 professional athletes from different sport sectors must undergo regular doping controls. As is the case with many other nations, Germany relies upon a system recommended by the World Anti-Doping Agency (WADA) to coordinate and initiate these controls. Known as the Anti-Doping Admission and Administration System, or ADAMS for short, the system requires athletes to submit their future regular whereabouts via browser, up to three months in advance. This allows the Doping Control Officer (DCO) to be able to locate the athlete without prior notice. The flip side, however, is that, since its introduction in 2007, ADAMS has not only been severely criticized by athletes themselves for its lack of usability, but has also come under attack by data-privacy experts accusing this system of breaching multiple data privacy laws.     

The correntropy MACE filter

The minimum average correlation energy (MACE) filter is well known for object recognition. This paper proposes a nonlinear extension to the MACE filter using the recently introduced correntropy function. Correntropy is a positive definite function that generalizes the concept of correlation by utilizing second and higher order moments of the signal statistics. Because of its positive definite nature, correntropy induces a new reproducing kernel Hilbert space (RKHS). Taking advantage of the linear structure of the RKHS it is possible to formulate the MACE filter equations in the RKHS induced by correntropy and obtained an approximate solution. Due to the nonlinear relation between the feature space and the input space, the correntropy MACE (CMACE) can potentially improve upon the MACE performance while preserving the shift-invariant property (additional computation for all shifts will be required in the CMACE). To alleviate the computation complexity of the solution, this paper also presents the fast CMACE using the fast Gauss transform (FGT). We apply the CMACE filter to the MSTAR public release synthetic aperture radar (SAR) data set as well as PIE database of human faces and show that the proposed method exhibits better distortion tolerance and outperforms the linear MACE in both generalization and rejection abilities.     

Hybrid systems: from verification to falsification by combining motion planning and discrete search

We propose HyDICE, Hybrid Discrete Continuous Exploration, a multi-layered approach for hybrid-system falsification that combines motion planning with discrete search and discovers safety violations by computing witness trajectories to unsafe states. The discrete search uses discrete transitions and a state-space decomposition to guide the motion planner during the search for witness trajectories. Experiments on a nonlinear hybrid robotic system with over one million modes and experiments with an aircraft conflict-resolution protocol with high-dimensional continuous state spaces demonstrate the effectiveness of HyDICE. Comparisons to related work show computational speedups of up to two orders of magnitude. Work supported in part by NSF CNS 0615328 (EP, LK, MV), NSF 0713623 (EP, LK), a Sloan Fellowship (LK), and NSF CCF 0613889 (MV). Equipment supported by NSF CNS 0454333 and NSF CNS 0421109 in partnership with Rice, AMD, and Cray. A preliminary version of this work was published in the Proceedings of the 19th International Conference on Computer Aided Verification (CAV 2007). Lecture Notes in Computer Science, eds. W. Damm and H. Hermanns, vol. 4590, pp. 468–481. This work contains substantial improvements to the overall computational method introduced in the preliminary work and new experiments that were not included in the preliminary work.     

Using EMG to anticipate head motion for virtual-environment applications

In virtual environment (VE) applications, where virtual objects are presented in a see-through head-mounted display, virtual images must be continuously stabilized in space in response to user's head motion. Time delays in head-motion compensation cause virtual objects to "swim" around instead of being stable in space which results in misalignment errors when overlaying virtual and real objects. Visual update delays are a critical technical obstacle for implementing head-mounted displays in applications such as battlefield simulation/training, telerobotics, and telemedicine. Head motion is currently measurable by a head-mounted 6-degrees-of-freedom inertial measurement unit. However, even given this information, overall VE-system latencies cannot be reduced under about 25 ms. We present a novel approach to eliminating latencies, which is premised on the fact that myoelectric signals from a muscle precede its exertion of force, thereby limb or head acceleration. We thus suggest utilizing neck-muscles' myoelectric signals to anticipate head motion. We trained a neural network to map such signals onto equivalent time-advanced inertial outputs. The resulting network can achieve time advances of up to 70 ms.