TIPICAL -- Type Inference for Python In Critical Accuracy Level

• URL: http://arxiv.org/abs/2308.02675v1
• Date: Fri, 4 Aug 2023 19:16:23 GMT
• Title: TIPICAL -- Type Inference for Python In Critical Accuracy Level
• Authors: Jonathan Elkobi, Bernd Gruner, Tim Sonnekalb, Clemens-Alexander Brust
• Abstract summary: TIPICAL is a method that combines deep similarity learning with novelty detection. We show that our method can better predict data types in high confidence by successfully filtering out unknown and inaccurate predicted data types.
• Score: 1.1666234644810896
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Type inference methods based on deep learning are becoming increasingly popular as they aim to compensate for the drawbacks of static and dynamic analysis approaches, such as high uncertainty. However, their practical application is still debatable due to several intrinsic issues such as code from different software domains will involve data types that are unknown to the type inference system. In order to overcome these problems and gain high-confidence predictions, we thus present TIPICAL, a method that combines deep similarity learning with novelty detection. We show that our method can better predict data types in high confidence by successfully filtering out unknown and inaccurate predicted data types and achieving higher F1 scores to the state-of-the-art type inference method Type4Py. Additionally, we investigate how different software domains and data type frequencies may affect the results of our method.

Related papers

• Diverse Data Augmentation with Diffusions for Effective Test-time Prompt Tuning [73.75282761503581]
  We propose DiffTPT, which leverages pre-trained diffusion models to generate diverse and informative new data. Our experiments on test datasets with distribution shifts and unseen categories demonstrate that DiffTPT improves the zero-shot accuracy by an average of 5.13%.
  arXiv  Detail & Related papers  (2023-08-11T09:36:31Z)
• Comprehensive Assessment of the Performance of Deep Learning Classifiers Reveals a Surprising Lack of Robustness [2.1320960069210484]
  This article advocates bench-marking performance using a wide range of different types of data. It is found that current deep neural networks, including those trained with methods that are believed to produce state-of-the-art robustness, are extremely vulnerable to making mistakes on certain types of data.
  arXiv  Detail & Related papers  (2023-08-08T08:50:27Z)
• Adaptive Negative Evidential Deep Learning for Open-set Semi-supervised Learning [69.81438976273866]
  Open-set semi-supervised learning (Open-set SSL) considers a more practical scenario, where unlabeled data and test data contain new categories (outliers) not observed in labeled data (inliers) We introduce evidential deep learning (EDL) as an outlier detector to quantify different types of uncertainty, and design different uncertainty metrics for self-training and inference. We propose a novel adaptive negative optimization strategy, making EDL more tailored to the unlabeled dataset containing both inliers and outliers.
  arXiv  Detail & Related papers  (2023-03-21T09:07:15Z)
• Benchmarking common uncertainty estimation methods with histopathological images under domain shift and label noise [62.997667081978825]
  In high-risk environments, deep learning models need to be able to judge their uncertainty and reject inputs when there is a significant chance of misclassification. We conduct a rigorous evaluation of the most commonly used uncertainty and robustness methods for the classification of Whole Slide Images. We observe that ensembles of methods generally lead to better uncertainty estimates as well as an increased robustness towards domain shifts and label noise.
  arXiv  Detail & Related papers  (2023-01-03T11:34:36Z)
• Cross-Domain Evaluation of a Deep Learning-Based Type Inference System [0.44098366957385177]
  We investigate Type4Py as a representative of state-of-the-art deep learning-based type inference systems. We address the following problems: class imbalances, out-of-vocabulary words, dataset shifts, and unknown classes. Our dataset enables the evaluation of type inference systems in different domains of software projects.
  arXiv  Detail & Related papers  (2022-08-19T07:28:31Z)
• Decorrelative Network Architecture for Robust Electrocardiogram Classification [4.808817930937323]
  It is not possible to train networks that are accurate in all scenarios. Deep learning methods sample the model parameter space to estimate uncertainty. These parameters are often subject to the same vulnerabilities, which can be exploited by adversarial attacks. We propose a novel ensemble approach based on feature decorrelation and Fourier partitioning for teaching networks diverse complementary features.
  arXiv  Detail & Related papers  (2022-07-19T02:36:36Z)
• Evaluating Membership Inference Through Adversarial Robustness [6.983991370116041]
  We propose an enhanced methodology for membership inference attacks based on adversarial robustness. We evaluate our proposed method on three datasets: Fashion-MNIST, CIFAR-10, and CIFAR-100.
  arXiv  Detail & Related papers  (2022-05-14T06:48:47Z)
• Improving robustness of jet tagging algorithms with adversarial training [56.79800815519762]
  We investigate the vulnerability of flavor tagging algorithms via application of adversarial attacks. We present an adversarial training strategy that mitigates the impact of such simulated attacks.
  arXiv  Detail & Related papers  (2022-03-25T19:57:19Z)
• Positive-Unlabeled Classification under Class-Prior Shift: A Prior-invariant Approach Based on Density Ratio Estimation [85.75352990739154]
  We propose a novel PU classification method based on density ratio estimation. A notable advantage of our proposed method is that it does not require the class-priors in the training phase.
  arXiv  Detail & Related papers  (2021-07-11T13:36:53Z)
• Detecting unusual input to neural networks [0.48733623015338234]
  We study a method that judges the unusualness of an input by evaluating its informative content compared to the learned parameters. This technique can be used to judge whether a network is suitable for processing a certain input and to raise a red flag that unexpected behavior might lie ahead.
  arXiv  Detail & Related papers  (2020-06-15T10:48:43Z)
• Meta-Learned Confidence for Few-shot Learning [60.6086305523402]
  A popular transductive inference technique for few-shot metric-based approaches, is to update the prototype of each class with the mean of the most confident query examples. We propose to meta-learn the confidence for each query sample, to assign optimal weights to unlabeled queries. We validate our few-shot learning model with meta-learned confidence on four benchmark datasets.
  arXiv  Detail & Related papers  (2020-02-27T10:22:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.