Jointly learning to localize and repair variable-misuse bugs improves program repair.
problem Variable-misuse bugs in programs.
method Multi-headed pointer networks for joint localization and repair.
result Joint model significantly outperforms an enumerative solution.
SequenceR uses seq-to-seq learning to fix bugs in code.
problem Fixing bugs in code using machine learning.
method Sequence-to-sequence learning with copy mechanism, trained on curated code samples.
result SequenceR can perfectly predict and find correct patches for bugs.
Neural network identifies undeclared variables and infers their types.
problem Undeclared variable errors in programs.
method Trained on structural semantic details of AST, identifies and infers types of undeclared variables.
result Correctly identified and inferred types for 80% of programs with undeclared variable errors.
Develops deep learning for logical code segmentation.
problem Lack of logically segmented source code.
method Novel deep learning approach to generate logical code segments.
result Improves software analysis tasks like commenting, bug detection, and code synthesis.
We explore the perspective of a bug living on the two-dimensional surface of a polyhedron. Images of various kinds of effects like lensing and cloaking are shown via color pictures of three viewpoints: the first person perspective of the bug, a map of the bug's viewpoint, and a look at the bug on the embedded polyhedro…
In this study we model the warranty claims process and evaluate the warranty servicing costs under non-renewing and renewing free repair warranties. We assume that the repair time for rectifying the claims is non-zero and the repair cost is a function of the length of the repair time. To accommodate the ageing of the p…
Graph-based approach repairs programs from diagnostic feedback.
problem Learning to repair programs from limited labeled data and compiler error messages.
method Introduces program-feedback graph and graph neural network for reasoning, and self-supervised learning with unlabeled programs.
result DrRepair significantly outperforms prior work, achieving high repair rates.
MACER accelerates error repair by modularly identifying and applying fixes.
problem Automated compilation error repair for novice programmers.
method Modular segregation of repair process into identification and application, using discriminative learning techniques.
result MACER outperforms existing methods by 20% on popular errors and is competitive on all error types.
Develops a deep metric learning approach for detecting bugs in video games.
problem Automated detection of bugs in video games.
method State-State Siamese Networks (S3N) for deep metric learning.
result S3N learns meaningful embeddings to identify various types of bugs.
Training set bugs are flaws in the data that adversely affect machine learning. The training set is usually too large for man- ual inspection, but one may have the resources to verify a few trusted items. The set of trusted items may not by itself be adequate for learning, so we propose an algorithm that uses these ite…
New method to recover over-parameterized models corrupted during estimation.
problem Recovering statistical models corrupted after initial estimation.
method Robust estimation using over-parameterized models and redundancy.
result Stochastic gradient descent is well-suited for model repair, but sparsity is generally not repairable.
RVAE detects and repairs corrupted cells in mixed-type tabular data.
problem Outlier detection and repair in mixed-type tabular data.
method Robust Variational Autoencoder (RVAE) learns the joint distribution of clean data and identifies outlier cells.
result RVAE outperforms state-of-the-art methods in cell outlier detection and repair for tabular data.
Proposes a method to repair arbitrage in option prices data.
problem Arbitrage in option price data can lead to poor performance or failure of financial applications.
method Formulates data repair as a linear programming (LP) problem to minimise price changes within bid and ask price bounds.
result The proposed method gives sparse perturbations on data and improves model calibration with enhanced robustness and reduced calibration error.
Motivated by the problem of automated repair of software vulnerabilities, we propose an adversarial learning approach that maps from one discrete source domain to another target domain without requiring paired labeled examples or source and target domains to be bijections. We demonstrate that the proposed adversarial l…
CLSVAE repairs systematic errors in images with minimal labeled data.
problem Repairing systematic errors in data, especially in images.
method CLSVAE models inliers as a smaller latent space representation, separating inlier and outlier patterns.
result CLSVAE achieves superior repairs with less than 2% labeled data, outperforming other methods.
REPAIR mitigates variance collapse to enable linear interpolation between SGD solutions.
problem Linear interpolation between SGD solutions is difficult due to variance collapse in permuted activations.
method REPAIR rescales preactivations of interpolated networks to mitigate variance collapse.
result 60%-100% relative barrier reduction across various architectures and tasks.
We explore solutions for automated labeling of content in bug trackers and customer support systems. In order to do that, we classify content in terms of several criteria, such as priority or product area. In the first part of the paper, we provide an overview of existing methods used for text classification. These met…
SnareNet adds repair layers to neural networks to ensure outputs meet physical constraints.
problem Unconstrained neural network predictions violate physical or safety requirements.
method SnareNet appends a differentiable repair layer that navigates constraints to produce feasible outputs.
result SnareNet consistently improves objective quality while satisfying constraints more reliably.
Develops Kleinian Sphere Packings and Bugs, proving their arithmetic origins.
problem Understanding sphere packings and their arithmetic origins in various dimensions.
method Introduces Kleinian Sphere Packings and Bugs, extending Arithmeticity Theorem.
result Kleinian packings and Bugs come from Q-arithmetic lattices of simplest type.
LLMs excel at summarizing and repairing complex models without needing full models.
problem Understanding and repairing complex models like GAMs.
method Hierarchical reasoning and extensive background knowledge.
result LLMs can detect anomalies, describe reasons, and suggest repairs.
CROP verifies clean prefixes in reasoning traces, improving downstream repair accuracy.
problem Uncertainty in reasoning traces prevents full certification of entire responses.
method CROP selects a calibrated threshold to certify the longest prefix with low risk proxies.
result CROP improves downstream repair accuracy by preserving valid reasoning and discarding misleading suffixes.
CASP improves portfolio optimization by considering asset covariance.
problem Infeasibility in cardinality-constrained portfolio optimization.
method CASP uses volatility-normalized selection and covariance-aware projection.
result CASP-Basic delivers lower portfolio variance than standard Euclidean repair.
We present a new application and covering number bound for the framework of "Machine Learning with Operational Costs (MLOC)," which is an exploratory form of decision theory. The MLOC framework incorporates knowledge about how a predictive model will be used for a subsequent task, thus combining machine learning with t…
Generative model learns diverse fixes for program errors.
problem Uncertainty in predicting a single fix for program errors.
method Deep conditional variational autoencoder with diversity regularizer.
result Strong improvements over state-of-the-art approaches, generating up to 45% diverse fixes.
Deep reinforcement learning has led to several recent breakthroughs, though the learned policies are often based on black-box neural networks. This makes them difficult to interpret and to impose desired specification constraints during learning. We present an iterative framework, MORL, for improving the learned polici…
Graph neural network predicts JavaScript types with high accuracy.
problem Automatic code repair for JavaScript programs.
method Graph Neural Network model for token type prediction.
result Achieved above 90% accuracy in token type predictions.
The above named paper has been withdrawn. A colleague has observed a gap in the proof of isotopy invariance, which can be repaired by reducing the coefficients (which lie in (1/6)Z) of the antisymmetric kanji with chords incident with more than one component modulo 8Z. An analogous issue arises in considering the effec…
This work uses image generation models to find vision model bugs.
problem Automatically discovering failures in vision models.
method Conditional text-to-image generation and captioning models.
result Demonstrated utility of large-scale generative models to find vision model bugs.
Many modern data-intensive computational problems either require, or benefit from distance or similarity data that adhere to a metric. The algorithms run faster or have better performance guarantees. Unfortunately, in real applications, the data are messy and values are noisy. The distances between the data points are …
Automated graphics testing detects novel corruptions without manual labeling.
problem Detecting novel visual corruptions in graphics unit testing without manual labeling.
method Reproduces driver bugs to generate corruptions and uses Multiple Instance Learning (MIL) methods.
result Significantly outperforms unsupervised methods and discovers novel corruptions.
Withdrawn May 2005. There is an error in the even-dimensional case of the proof in the April 2005 version. The hoped-for 4-dimensional applications are unlikely to survive the repairs.
Suppose M is a compact Riemannian manifold and p∈M an arbitrary point. We employ estimates on the volume growth around p to prove that the only conformal compactification of M∖{p} is M itself.
We study graphs of (generalized) joins and intersections of finitely generated subgroups of a free group. We show how to disprove a lemma of Imrich and Müller on these graphs and how to repair this lemma.
TFCheck detects training issues in ML programs using TensorFlow.
problem Training programs often contain inconsistencies and bugs.
method Developed a TensorFlow library (TFCheck) with verification routines.
result TFCheck successfully detects training issues in ML code.
Bayes-optimal classifiers are robust to adversarial attacks, unlike CNNs trained on the same data.
problem The vulnerability of modern CNN classifiers to adversarial examples.
method Constructing realistic image datasets and deriving analytic conditions for Bayes-optimal classifiers.
result Bayes-optimal classifiers are robust to adversarial attacks, unlike CNNs trained on the same data.
This research simplifies verification of machine learning systems using reparameterization.
problem Reduce or eliminate serious bugs in machine learning systems.
method Use proof assistants to construct machine-checked proofs of correctness, leveraging reparameterization to handle probabilistic claims.
result Demonstrates broad applicability of reparameterization to verify different types of machine learning systems.
The paper develops methods to identify and correct buggy data in linear regression models.
problem Identifying and correcting buggy data in linear regression models.
method Formulated a general statistical algorithm for identifying buggy points, provided theoretical guarantees, and proposed an algorithm for tuning parameter selection.
result Theoretical guarantees and empirical results show the effectiveness of the proposed debugging algorithm.
SED integrates synthesis, execution, and debugging for neural program synthesis.
problem Challenges in synthesizing complex programs that match specifications.
method SED combines synthesis, execution, and debugging to improve neural program generation.
result SED reduces error rates and outperforms standard decoding methods.
A complete error analysis of variational integrators is obtained, by blowing up the discrete variational principles, all of which have a singularity at zero time-step. Divisions by the time step lead to an order that is one less than observed in simulations, a deficit that is repaired with the help of a new past-future…
The paper addresses fairness in machine learning by adjusting input distributions.
problem Reducing disparate impact in machine learning models over different groups.
method The approach involves learning a counterfactual distribution to adjust input variables for disadvantaged groups.
result The method can reduce disparate impact without training a new model.
Unified platform SOCRATES for neural network analysis.
problem Analyzing neural networks for bugs and fairness.
method Standardized format, assertion language, and multiple analysis algorithms.
result Unified platform for neural network analysis.
Algorithms learned from data are increasingly used for deciding many aspects in our life: from movies we see, to prices we pay, or medicine we get. Yet there is growing evidence that decision making by inappropriately trained algorithms may unintentionally discriminate people. For example, in automated matching of cand…
Forward Automatic Differentiation (AD) is a technique for augmenting programs to compute derivatives. The essence of Forward AD is to attach perturbations to each number, and propagate these through the computation. When derivatives are nested, the distinct derivative calculations, and their associated perturbations, m…
This paper addresses the problem of predicting duration of unplanned power outages, using historical outage records to train a series of neural network predictors. The initial duration prediction is made based on environmental factors, and it is updated based on incoming field reports using natural language processing …
Adversarial attacks found to be effective on code models.
problem Adversarial robustness of code models.
method Instantiated adversarial attacks for code, demonstrated vulnerability, and improved robustness.
result Neural models for code are vulnerable to adversarial attacks.
Expert-augmented algorithm boosts scores in Montezuma's Revenge.
problem Sparse rewards in Montezuma's Revenge.
method Expert-augmented actor-critic algorithm.
result Achieves above 27,000 points consistently, surpassing expert performance.
Optimal pre-processing reduces disparate impact by minimizing total variation distance.
problem Achieving fairness in data outputs based on protected attributes.
method Using pre-processing to enforce fairness, minimizing total variation distance between pre-processed and original data distributions.
result The problem of fairness can be formulated as a linear program, efficiently solvable.
Corrected CBOW performs similarly to Skip-gram.
problem CBOW embeddings underperform Skip-gram embeddings in word2vec.
method Fixed a bug in CBOW gradient update to improve performance.
result Corrected CBOW embeddings are competitive with Skip-gram on various tasks.