Clemens-Alexander Brust
YOU?
Author Swipe
View article: TIPICAL -- Type Inference for Python In Critical Accuracy Level
TIPICAL -- Type Inference for Python In Critical Accuracy Level Open
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 …
View article: A Static Analysis Platform for Investigating Security Trends in Repositories
A Static Analysis Platform for Investigating Security Trends in Repositories Open
Static analysis tools come in many forms andconfigurations, allowing them to handle various tasks in a (secure) development process: code style linting, bug/vulnerability detection, verification, etc., and adapt to the specific requirement…
View article: Generalizability of Code Clone Detection on CodeBERT
Generalizability of Code Clone Detection on CodeBERT Open
Transformer networks such as CodeBERT already achieve outstanding results for code clone detection in benchmark datasets, so one could assume that this task has already been solved. However, code clone detection is not a trivial task. Sema…
View article: Generalizability of Code Clone Detection on CodeBERT
Generalizability of Code Clone Detection on CodeBERT Open
Transformer networks such as CodeBERT already achieve outstanding results for code clone detection in benchmark datasets, so one could assume that this task has already been solved. However, code clone detection is not a trivial task. Sema…
View article: Cross-Domain Evaluation of a Deep Learning-Based Type Inference System
Cross-Domain Evaluation of a Deep Learning-Based Type Inference System Open
Optional type annotations allow for enriching dynamic programming languages with static typing features like better Integrated Development Environment (IDE) support, more precise program analysis, and early detection and prevention of type…
View article: Pre-trained models are not enough: active and lifelong learning is important for long-term visual monitoring of mammals in biodiversity research—Individual identification and attribute prediction with image features from deep neural networks and decoupled decision models applied to elephants and great apes
Pre-trained models are not enough: active and lifelong learning is important for long-term visual monitoring of mammals in biodiversity research—Individual identification and attribute prediction with image features from deep neural networks and decoupled decision models applied to elephants and great apes Open
Animal re-identification based on image data, either recorded manually by photographers or automatically with camera traps, is an important task for ecological studies about biodiversity and conservation that can be highly automatized with…
View article: CrossDomainTypes4Py: A Python Dataset for Cross-Domain Evaluation of Type Inference Systems
CrossDomainTypes4Py: A Python Dataset for Cross-Domain Evaluation of Type Inference Systems Open
This dataset contains python repositories mined on GitHub on January 20, 2021. It allows a cross-domain evaluation of type inference systems. For this purpose, it consists of two sub-datasets, each containing only projects from the web or …
View article: CrossDomainTypes4Py: A Python Dataset for Cross-Domain Evaluation of Type Inference Systems
CrossDomainTypes4Py: A Python Dataset for Cross-Domain Evaluation of Type Inference Systems Open
This dataset contains python repositories mined on GitHub on January 20, 2021. It allows a cross-domain evaluation of type inference systems. For this purpose, it consists of two sub-datasets, each containing only projects from the web or …
View article: Self-supervised Learning from Semantically Imprecise Data
Self-supervised Learning from Semantically Imprecise Data Open
Learning from imprecise labels such as animal or bird, but making precise predictions like snow bunting at inference time is an important capability for any classifier when expertly labeled training data is scarce. Contributions by volunte…
View article: ROMEO: Exploring Juliet through the Lens of Assembly Language
ROMEO: Exploring Juliet through the Lens of Assembly Language Open
Automatic vulnerability detection on C/C++ source code has benefitted from the introduction of machine learning to the field, with many recent publications considering this combination. In contrast, assembly language or machine code artifa…
View article: ROMEO: Exploring Juliet through the Lens of Assembly Language
ROMEO: Exploring Juliet through the Lens of Assembly Language Open
Automatic vulnerability detection on C/C++ source code has benefitted from the introduction of machine learning to the field, with many recent publications targeting this combination. In contrast, assembly language or machine code artifact…
View article: Self-Supervised Learning from Semantically Imprecise Data
Self-Supervised Learning from Semantically Imprecise Data Open
Learning from imprecise labels such as "animal" or "bird", but making precise predictions like "snow bunting" at inference time is an important capability for any classifier when expertly labeled training data is scarce. Contributions by v…
View article: Making Every Label Count: Handling Semantic Imprecision by Integrating Domain Knowledge
Making Every Label Count: Handling Semantic Imprecision by Integrating Domain Knowledge Open
Noisy data, crawled from the web or supplied by volunteers such as Mechanical Turkers or citizen scientists, is considered an alternative to professionally labeled data. There has been research focused on mitigating the effects of label no…
View article: Making Every Label Count: Handling Semantic Imprecision by Integrating\n Domain Knowledge
Making Every Label Count: Handling Semantic Imprecision by Integrating\n Domain Knowledge Open
Noisy data, crawled from the web or supplied by volunteers such as Mechanical\nTurkers or citizen scientists, is considered an alternative to professionally\nlabeled data. There has been research focused on mitigating the effects of\nlabel…
View article: Active Learning for Deep Object Detection
Active Learning for Deep Object Detection Open
The great success that deep models have achieved in the past is mainly owed to large amounts of labeled training data. However, the acquisition of labeled data for new tasks aside from existing benchmarks is both challenging and costly. Ac…
View article: Not just a matter of semantics: the relationship between visual similarity and semantic similarity
Not just a matter of semantics: the relationship between visual similarity and semantic similarity Open
Knowledge transfer, zero-shot learning and semantic image retrieval are methods that aim at improving accuracy by utilizing semantic information, e.g. from WordNet. It is assumed that this information can augment or replace missing visual …
View article: Integrating domain knowledge: using hierarchies to improve deep\n classifiers
Integrating domain knowledge: using hierarchies to improve deep\n classifiers Open
One of the most prominent problems in machine learning in the age of deep\nlearning is the availability of sufficiently large annotated datasets. For\nspecific domains, e.g. animal species, a long-tail distribution means that some\nclasses…
View article: Active Learning for Deep Object Detection
Active Learning for Deep Object Detection Open
The great success that deep models have achieved in the past is mainly owed to large amounts of labeled training data. However, the acquisition of labeled data for new tasks aside from existing benchmarks is both challenging and costly. Ac…
View article: Keeping the Human in the Loop: Towards Automatic Visual Monitoring in Biodiversity Research
Keeping the Human in the Loop: Towards Automatic Visual Monitoring in Biodiversity Research Open
More and more methods in the area of biodiversity research grounds upon new opportunities arising from modern sensing devices that in principle make it possible to continuously record sensor data from the environment. However, these opport…
View article: Neither Quick Nor Proper -- Evaluation of QuickProp for Learning Deep Neural Networks
Neither Quick Nor Proper -- Evaluation of QuickProp for Learning Deep Neural Networks Open
Neural networks and especially convolutional neural networks are of great interest in current computer vision research. However, many techniques, extensions, and modifications have been published in the past, which are not yet used by curr…
View article: Evaluation of QuickProp for Learning Deep Neural Networks -- A Critical Review
Evaluation of QuickProp for Learning Deep Neural Networks -- A Critical Review Open
Neural networks and especially convolutional neural networks are of great interest in current computer vision research. However, many techniques, extensions, and modifications have been published in the past, which are not yet used by curr…
View article: Convolutional Patch Networks with Spatial Prior for Road Detection and Urban Scene Understanding
Convolutional Patch Networks with Spatial Prior for Road Detection and Urban Scene Understanding Open
Classifying single image patches is important in many different applications, such as road detection or scene understanding. In this paper, we present convolutional patch networks, which are convolutional networks learned to distinguish di…