Subtle Errors Research Articles

Just-in-time static type checking for dynamic languages

Dynamic languages such as Ruby, Python, and JavaScript have many compelling benefits, but the lack of static types means subtle errors can remain latent in code for a long time. While many researchers have developed various systems to bring some of the benefits of static types to dynamic languages, prior approaches have trouble dealing with metaprogramming, which generates code as the program executes. In this paper, we propose Hummingbird, a new system that uses a novel technique, just-in-time static type checking, to type check Ruby code even in the presence of metaprogramming. In Hummingbird, method type signatures are gathered dynamically at run-time, as those methods are created. When a method is called, Hummingbird statically type checks the method body against current type signatures. Thus, Hummingbird provides thorough static checks on a per-method basis, while also allowing arbitrarily complex metaprogramming. For performance, Hummingbird memoizes the static type checking pass, invalidating cached checks only if necessary. We formalize Hummingbird using a core, Ruby-like language and prove it sound. To evaluate Hummingbird, we applied it to six apps, including three that use Ruby on Rails, a powerful framework that relies heavily on metaprogramming. We found that all apps typecheck successfully using Hummingbird, and that Hummingbird's performance overhead is reasonable. We applied Hummingbird to earlier versions of one Rails app and found several type errors that had been introduced and then fixed. Lastly, we demonstrate using Hummingbird in Rails development mode to typecheck an app as live updates are applied to it.

On-Chip Cube-Based Constrained-Random Stimuli Generation for Post-Silicon Validation

Post-silicon validation is critical for exposing subtle design errors that have escaped to the silicon prototypes. Its effectiveness is conditioned by in-system application of a large volume of functionally-compliant stimuli. In this paper, we present a methodology to design constrained-random stimuli generators, which are placed on-chip and are configurable at design-time to generate in-system functionally-compliant stimuli subject to user-programmable constraints provided at validation time. Central to our method is a cube-based representation of constraints. These cubes are used as masks that force pseudo-random sequences to map onto functionally-compliant stimuli. To reduce the on-chip storage requirements, masks are compressed at design-time and expanded on-the-fly at validation time using decompression circuitry. Experimental results evaluate the impact of our method on the requirements for on-chip logic and memory resources.

Words are not enough: nonword repetition as an indicator of arcuate fasciculus integrity during brain tumor resection

OBJECTIVE Subcortical electrical stimulation during brain surgery may allow localization of functionally crucial white matter fibers and thus tailoring of the tumor resection according to its functional limits. The arcuate fasciculus (AF) is a white matter bundle connecting frontal, temporal, and parietal cortical areas that is often disrupted by left brain lesions. It plays a critical role in several cognitive functions related to phonological processing, but current intraoperative monitoring methods do not yet allow mapping of this tract with sufficient precision. In the present study the authors aimed to test a new paradigm for the intraoperative monitoring of the AF. METHODS In this report, the authors studied 12 patients undergoing awake brain surgery for tumor resection with a related risk of AF damage. To preserve AF integrity and the cognitive processes sustained by this tract in the intraoperative context, the authors used real word repetition (WR) and nonword repetition (NWR) tasks as complements to standard picture naming. RESULTS Compared with the errors identified by WR or picture naming, the NWR task allowed the detection of subtle errors possibly related to AF alterations. Moreover, only 3 patients demonstrated phonological paraphasias in standard picture naming, and in 2 of these patients the paraphasias co-occurred with the total loss of WR and NWR ability. Before surgery, lesion volume predicted a patient's NWR performance. CONCLUSIONS The authors suggest that monitoring NWR intraoperatively may complement the standard naming tasks and could permit better preservation of the important language production functions subserved by the AF.

Developmental differences in beta and theta power during sentence processing

Although very young children process ongoing language quickly and effortlessly, research indicates that they continue to improve and mature in their language skills through adolescence. This prolonged development may be related to differing engagement of semantic and syntactic processes. This study used event related potentials and time frequency analysis of EEG to identify developmental differences in neural engagement as children (ages 10–12) and adults performed an auditory verb agreement grammaticality judgment task. Adults and children revealed very few differences in comprehending grammatically correct sentences. When identifying grammatical errors, however, adults displayed widely distributed beta and theta power decreases that were significantly less pronounced in children. Adults also demonstrated a significant P600 effect, while children exhibited an apparent N400 effect. Thus, when identifying subtle grammatical errors in real time, adults display greater neural activation that is traditionally associated with syntactic processing whereas children exhibit greater activity more commonly associated with semantic processing. These findings support previous claims that the cognitive and neural underpinnings of syntactic processing are still developing in adolescence, and add to them by more clearly identifying developmental changes in the neural oscillations underlying grammatical processing.

Automated Selection of Assertions for Bit-Flip Detection During Post-Silicon Validation

Post-silicon validation deals with detection and diagnosis of errors that, due to existing limitations in pre-silicon verification, escape to the silicon prototypes and need to be fixed before committing to high-volume manufacturing. Electrical errors, such as those caused by cross-talk or power droops, are particularly difficult to catch during the pre-silicon phase because of the insufficient accuracy of device models, which is often traded-off against simulation time. This challenge is further aggravated by the rising number of voltage domains, especially if subtle errors are excited in unique electrical states. In fact these electrically-induced subtle errors most commonly manifest in the logic domain as bit-flips and, to the best of our knowledge, there are no systematic methods for designing embedded hardware monitors for generic logic blocks that can detect bit-flips with low detection latency. Moreover, unlike pre-silicon verification and manufacturing test that benefit from well-defined and universally accepted coverage metrics, there is no generic metric from which confidence can be implied at the end of post-silicon validation. Toward these goals, we present a method that relies on design invariants (assertions) that are ranked based on their potential to detect bit-flips. We also introduce two metrics bit-flip coverage estimate and flip-flop coverage estimate that can be used to assess the quality of the selected assertions, and, in general, the effectiveness of the post-silicon validation process.

GIAO C-H COSY Simulations Merged with Artificial Neural Networks Pattern Recognition Analysis. Pushing the Structural Validation a Step Forward.

The structural validation problem using quantum chemistry approaches (confirm or reject a candidate structure) has been tackled with artificial neural network (ANN) mediated multidimensional pattern recognition from experimental and calculated 2D C-H COSY. In order to identify subtle errors (such as regio- or stereochemical), more than 400 ANNs have been built and trained, and the most efficient in terms of classification ability were successfully validated in challenging real examples of natural product misassignments.

Commentary on decision-making and judgments in child maltreatment prevention and response: An overview

Professional judgments and decisions in alleged cases of child abuse and neglect can be difficult and complex, often based on ambiguous, partial, and contradictory information. These decisions have potentially fateful consequences not only for the child, but also for the entire family. Yet, decision-making in Child Protective Services (CPS) is characterized by low reliability and frequent errors. Examples abound of apparent errors where CPS did not act when it should have, or acted when it should not have. Aside from highly visible errors, other more subtle decision errors can occur that result in overall outcomes that are less than desirable for children and families. To improve decision-making, CPS decision makers at all levels can benefit from taking advantage of the knowledge gains and progress regarding decision-making research. The intent of this issue is to present research that articulates and explains how context and decision-maker behaviors effect CPS decision-making and how such knowledge might lead to improvements in decision-making.... Language: en

Rampant software errors may undermine scientific results.

The opportunities for both subtle and profound errors in software and data management are boundless, yet they remain surprisingly underappreciated. Here I estimate that any reported scientific result could very well be wrong if data have passed through a computer, and that these errors may remain largely undetected. It is therefore necessary to greatly expand our efforts to validate scientific software and computed results.

Rampant software errors may undermine scientific results

The opportunities for both subtle and profound errors in software and data management are boundless, yet they remain surprisingly underappreciated. Here I estimate that any reported scientific result could very well be wrong if data have passed through a computer, and that these errors may remain largely undetected. It is therefore necessary to greatly expand our efforts to validate scientific software and computed results.

Synthesizing racy tests

Subtle concurrency errors in multithreaded libraries that arise because of incorrect or inadequate synchronization are often difficult to pinpoint precisely using only static techniques. On the other hand, the effectiveness of dynamic race detectors is critically dependent on multithreaded test suites whose execution can be used to identify and trigger races. Usually, such multithreaded tests need to invoke a specific combination of methods with objects involved in the invocations being shared appropriately to expose a race. Without a priori knowledge of the race, construction of such tests can be challenging. In this paper, we present a lightweight and scalable technique for synthesizing precisely these kinds of tests. Given a multithreaded library and a sequential test suite, we describe a fully automated analysis that examines sequential execution traces, and produces as its output a concurrent client program that drives shared objects via library method calls to states conducive for triggering a race. Experimental results on a variety of well-tested Java libraries yield 101 synthesized multithreaded tests in less than four minutes. Analyzing the execution of these tests using an off-the-shelf race detector reveals 187 harmful races, including several previously unreported ones. Our implementation, named NARADA, and the results of our experiments are available at http://www.csa.iisc.ernet.in/~sss/tools/narada.

Rampant software errors undermine scientific results

Errors in scientific results due to software bugs are not limited to a few high-profile cases that lead to retractions and are widely reported. Here I estimate that in fact most scientific results are probably wrong if data have passed through a computer, and that these errors may remain largely undetected. The opportunities for both subtle and profound errors in software and data management are boundless, yet they remain surprisingly underappreciated.

Alignment Accuracy and Intrafraction Motion Analysis in Frameless Functional Image Guided Radiosurgery

Alignment Accuracy and Intrafraction Motion Analysis in Frameless Functional Image Guided Radiosurgery

A new screening method for recombinant Saccharomyces cerevisiae strains based on their xylose fermentation ability measured by near infrared spectroscopy

Fuel ethanol produced from lignocellulose by the yeast Saccharomyces cerevisiae becomes an increasingly important alternative to fossil fuels. Selection of S. cerevisiae strains, which can effectively produce ethanol from xylose is crucial to improve the fuel yield from lignocellulose. In the present study, a universal calibration system was developed by the combination of time series, fermentation near infrared (NIR) spectral data analysis and reference high-performance analysis of a single yeast strain which enabled the evaluation of the ethanol production ability of a wide variety of xylose-fermenting yeast strains. Subtraction of xylose and ethanol concentrations at 0 h for each clone, as well as the respective spectra, reduced subtle errors of the fermentation components naturally occurring in multiple experiments to clearly visualize the difference of fermentation ability between strains. Also, NIR spectra showed specific peaks in difference spectra calculated from the subtraction treatment. A robust univariate linear regression model led to high prediction accuracy of xylose consumption (R2 > 0.99) and ethanol production (R2 > 0.98) for a wide variety of yeast strains when using only the distinct spectral pattern of a single-strain. Here, a novel method for screening of high-performing yeast strains has been developed requiring only a simple single-strain calibration model. The advantages of NIR spectroscopy such as rapid and convenient experimental preparation using electromagnetic waves in the region, which provide deep penetration into an aqueous sample, are successfully exploited in the proposed screening method.

Race detection for Android applications

Programming environments for smartphones expose a concurrency model that combines multi-threading and asynchronous event-based dispatch. While this enables the development of efficient and feature-rich applications, unforeseen thread interleavings coupled with non-deterministic reorderings of asynchronous tasks can lead to subtle concurrency errors in the applications. In this paper, we formalize the concurrency semantics of the Android programming model. We further define the happens-before relation for Android applications, and develop a dynamic race detection technique based on this relation. Our relation generalizes the so far independently studied happens-before relations for multi-threaded programs and single-threaded event-driven programs. Additionally, our race detection technique uses a model of the Android runtime environment to reduce false positives. We have implemented a tool called DroidRacer. It generates execution traces by systematically testing Android applications and detects data races by computing the happens-before relation on the traces. We analyzed 15 Android applications including popular applications such as Facebook, Twitter and K-9 Mail. Our results indicate that data races are prevalent in Android applications, and that DroidRacer is an effective tool to identify data races.

TU‐C‐BRE‐05: Clinical Implications of AAA Commissioning Errors and Ability of Common Commissioning ' Credentialing Procedures to Detect Them

Purpose:To test the ability of the TG‐119 commissioning process and RPC credentialing to detect errors in the commissioning process for a commercial Treatment Planning System (TPS).Methods:We introduced commissioning errors into the commissioning process for the Anisotropic Analytical Algorithm (AAA) within the Eclipse TPS. We included errors in Dosimetric Leaf Gap (DLG), electron contamination, flattening filter material, and beam profile measurement with an inappropriately large farmer chamber (simulated using sliding window smoothing of profiles). We then evaluated the clinical impact of these errors on clinical intensity modulated radiation therapy (IMRT) plans (head and neck, low and intermediate risk prostate, mesothelioma, and scalp) by looking at PTV D99, and mean and max OAR dose. Finally, for errors with substantial clinical impact we determined sensitivity of the RPC IMRT film analysis at the midpoint between PTV and OAR using a 4mm distance to agreement metric, and of a 7% TLD dose comparison. We also determined sensitivity of the 3 dose planes of the TG‐119 C‐shape IMRT phantom using gamma criteria of 3% 3mm.Results:The largest clinical impact came from large changes in the DLG with a change of 1mm resulting in up to a 5% change in the primary PTV D99. This resulted in a discrepancy in the RPC TLDs in the PTVs and OARs of 7.1% and 13.6% respectively, which would have resulted in detection. While use of incorrect flattening filter caused only subtle errors (<1%) in clinical plans, the effect was most pronounced for the RPC TLDs in the OARs (>6%).Conclusion:The AAA commissioning process within the Eclipse TPS is surprisingly robust to user error. When errors do occur, the RPC and TG‐119 commissioning credentialing criteria are effective at detecting them; however OAR TLDs are the most sensitive despite the RPC currently excluding them from analysis.

SU-E-T-152: Error Sensitivity and Superiority of a Protocol for 3D IMRT Quality Assurance

Purpose: To test if the parameters included in our 3D QA protocol with current tolerance levels are able to detect certain errors and show the superiority of 3D QA method over single ion chamber measurements and 2D gamma test by detecting most of the introduced errors. The 3D QA protocol parameters are: TPS and measured average dose difference, 3D gamma test with 3mmDTA/3% test parameters, and structure volume for which the TPS predicted and measured absolute dose difference is greater than 6%. Methods: Two prostate and two thoracic step-and-shoot IMRT patients were investigated. The following errors were introduced to each original treatment plan: energy switched from 6MV to 10MV, linac jaws retracted to 15cmx15cm, 1,2,3 central MLC leaf pairs retracted behind the jaws, single central MLC leaf put in or out of the treatment field, Monitor Units (MU) increased and decreased by 1 and 3%, collimator off by 5 and 15 degrees, detector shifted by 5mm to the left and right, gantry treatment angle off by 5 and 15 degrees. QA was performed on each plan using single ion chamber, 2D ion chamber array for 2D gamma analysis and using IBA's COMPASS system for 3D QA. Results: Out of the three tested QA methods single ion chamber performs the worst not detecting subtle errors. 3D QA proves to be the superior out of the three methods detecting all of introduced errors, except 10MV and 1% MU change, and MLC rotated (those errors were not detected by any QA methods tested). Conclusion: As the way radiation is delivered evolves, so must the QA. We believe a diverse set of 3D statistical parameters applied both to OAR and target plan structures provides the highest level of QA.

REDUCING SYSTEMATIC ERROR IN WEAK LENSING CLUSTER SURVEYS

Weak lensing provides an important route toward collecting samples of clusters of galaxies selected by mass. Subtle systematic errors in image reduction can compromise the power of this technique. We use the B-mode signal to quantify this systematic error and to test methods for reducing this error. We show that two procedures are efficient in suppressing systematic error in the B-mode: (1) refinement of the mosaic CCD warping procedure to conform to absolute celestial coordinates and (2) truncation of the smoothing procedure on a scale of 10$^{\prime}$. Application of these procedures reduces the systematic error to 20% of its original amplitude. We provide an analytic expression for the distribution of the highest peaks in noise maps that can be used to estimate the fraction of false peaks in the weak lensing $\kappa$-S/N maps as a function of the detection threshold. Based on this analysis we select a threshold S/N = 4.56 for identifying an uncontaminated set of weak lensing peaks in two test fields covering a total area of $\sim 3$deg$^2$. Taken together these fields contain seven peaks above the threshold. Among these, six are probable systems of galaxies and one is a superposition. We confirm the reliability of these peaks with dense redshift surveys, x-ray and imaging observations. The systematic error reduction procedures we apply are general and can be applied to future large-area weak lensing surveys. Our high peak analysis suggests that with a S/N threshold of 4.5, there should be only 2.7 spurious weak lensing peaks even in an area of 1000 deg$^2$ where we expect $\sim$ 2000 peaks based on our Subaru fields.

The importance of accurate adiabatic interaction potentials for the correct description of electronically nonadiabatic vibrational energy transfer: a combined experimental and theoretical study of NO(v = 3) collisions with a Au(111) surface.

We present a combined experimental and theoretical study of NO(v = 3 → 3, 2, 1) scattering from a Au(111) surface at incidence translational energies ranging from 0.1 to 1.2 eV. Experimentally, molecular beam-surface scattering is combined with vibrational overtone pumping and quantum-state selective detection of the recoiling molecules. Theoretically, we employ a recently developed first-principles approach, which employs an Independent Electron Surface Hopping (IESH) algorithm to model the nonadiabatic dynamics on a Newns-Anderson Hamiltonian derived from density functional theory. This approach has been successful when compared to previously reported NO/Au scattering data. The experiments presented here show that vibrational relaxation probabilities increase with incidence energy of translation. The theoretical simulations incorrectly predict high relaxation probabilities at low incidence translational energy. We show that this behavior originates from trajectories exhibiting multiple bounces at the surface, associated with deeper penetration and favored (N-down) molecular orientation, resulting in a higher average number of electronic hops and thus stronger vibrational relaxation. The experimentally observed narrow angular distributions suggest that mainly single-bounce collisions are important. Restricting the simulations by selecting only single-bounce trajectories improves agreement with experiment. The multiple bounce artifacts discovered in this work are also present in simulations employing electronic friction and even for electronically adiabatic simulations, meaning they are not a direct result of the IESH algorithm. This work demonstrates how even subtle errors in the adiabatic interaction potential, especially those that influence the interaction time of the molecule with the surface, can lead to an incorrect description of electronically nonadiabatic vibrational energy transfer in molecule-surface collisions.

Layout-versus-schematic verification for superconductive integrated circuits

Thorough layout verification of superconducting integrated circuits goes beyond design rule checking and parameter value extraction. The former is used to verify adherence to process design rules, and the latter to determine component element values. Still, neither gives much warning against subtle layout errors that result in unintended parasitic elements, or a layout that does not reflect the original circuit topology. A specialized implementation for Cadence Virtuoso allows layout-versus-schematic (LVS) verification, but is limited to commercial software and in terms of usefulness. Parameter extraction software such as InductEx require the circuit topology to be provided as a netlist, and element values are extracted for this topology even if a mistake in the layout or the netlist results in a model mismatch. Here we present a free-standing LVS verification toolkit for superconductive integrated circuits, and discuss its implementation. It comprises layout-to-schematic conversion that creates a first-pass netlist, netlist simplification, netlist and schematic visualization for user feedback, and netlist comparison to determine if a layout agrees with an input schematic. We show results for gate-level layouts and how it is used with InductEx to perform automated parameter extraction for layout verification. We conclude that this work makes layout verification more efficient and minimizes layout mistakes.

A comparison of methods for automated motion correction of DCE-MRI perfusion datasets evaluated in terms of diagnostic accuracy: a CE-MARC sub-study

Background Automated mage registration in cardiac myocardial perfusion is a necessity before quantitative perfusion can be widely accepted in clinical practice. Increasingly complex motion correction algorithms are being developed to deal with cardiac motion. However, the impact of these improvements has not been evaluated in terms of the final clinical diagnosis. Advanced motion correction methods are associated with increased computational overhead and the potential of introducing subtle registration errors, which can be hard to detect and quantify. The aim of this study was to compare the performance of the various automated correction methods in terms of their impact on diagnostic accuracy.