240,299 Hits in 3.6 sec

Deep learning type inference

Vincent J. Hellendoorn, Christian Bird, Earl T. Barr, Miltiadis Allamanis
2018 Proceedings of the 2018 26th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering - ESEC/FSE 2018  
We propose DeepTyper, a deep learning model that understands which types naturally occur in certain contexts and relations and can provide type suggestions, which can often be verified by the type checker  ...  , even if it could not infer the type initially.  ...  We use this data to train DeepTyper, which uses deep learning on existing typed code to infer new type annotations for JS and TS.  ... 
doi:10.1145/3236024.3236051 dblp:conf/sigsoft/HellendoornBBA18 fatcat:nytas5cz3vbbhbfxqlgoxagj7i

Type4Py: Deep Similarity Learning-Based Type Inference for Python [article]

Amir M. Mir, Evaldas Latoskinas, Sebastian Proksch, Georgios Gousios
2021 arXiv   pre-print
In this paper, we present Type4Py, a deep similarity learning-based hierarchical neural network model that learns to discriminate between types of the same kind and dissimilar types in a high-dimensional  ...  As retrofitting types to existing codebases is error-prone and laborious, learning-based approaches have been proposed to enable automatic type annotations based on existing, partially annotated codebases  ...  Motivated by the above discussion, in this paper, we present a deep similarity learning (DSL)-based type inference approach, called Type4Py.  ... 
arXiv:2101.04470v2 fatcat:wdyshjbse5ba3o2akw3btluncy

Advanced Graph-Based Deep Learning for Probabilistic Type Inference [article]

Fangke Ye, Jisheng Zhao, Vivek Sarkar
2021 arXiv   pre-print
Previous work has demonstrated the promise of probabilistic type inference using deep learning.  ...  and 86.89% respectively, outperforming the two most closely related deep learning type inference approaches from past work -- DeepTyper with a top-1 accuracy of 84.62% and LambdaNet with a top-1 accuracy  ...  RQ.3 : How does our approach compare with state-of-the-art deep learning type inference approaches?  ... 
arXiv:2009.05949v2 fatcat:yajlk2zvonarvcp4o3kpv6tcze

AdaptivFloat: A Floating-point based Data Type for Resilient Deep Learning Inference [article]

Thierry Tambe, En-Yu Yang, Zishen Wan, Yuntian Deng, Vijay Janapa Reddi, Alexander Rush, David Brooks, Gu-Yeon Wei
2020 arXiv   pre-print
We present AdaptivFloat, a floating-point inspired number representation format for deep learning that dynamically maximizes and optimally clips its available dynamic range, at a layer granularity, in  ...  AdaptivFloat consistently produces higher inference accuracies compared to block floating-point, uniform, IEEE-like float or posit encodings at very low precision (≤ 8-bit) across a diverse set of state-of-the-art  ...  INTRODUCTION Deep learning approaches have transformed representation learning in a multitude of tasks.  ... 
arXiv:1909.13271v3 fatcat:64jhd7jt4bhytf77wv4ezyrd7y

PYInfer: Deep Learning Semantic Type Inference for Python Variables [article]

Siwei Cui, Gang Zhao, Zeyu Dai, Luochao Wang, Ruihong Huang, Jeff Huang
2021 arXiv   pre-print
In this paper, we propose PYInfer, an end-to-end learning-based type inference tool that automatically generates type annotations for Python variables.  ...  Python type inference is challenging in practice.  ...  We highlight that fusing deep learning with static analysis to infer type annotations is promising.  ... 
arXiv:2106.14316v1 fatcat:w2lfh2ouyfg6bemdeypph7s634

DLTPy: Deep Learning Type Inference of Python Function Signatures using Natural Language Context [article]

Casper Boone, Niels de Bruin, Arjan Langerak, Fabian Stelmach
2019 arXiv   pre-print
In this paper, we present DLTPy: a deep learning type inference solution for the prediction of types in function signatures based on the natural language context (identifier names, comments and return  ...  To have the benefits of static typing, combined with high developer productivity, types need to be inferred.  ...  We present DLTPy as a deep learning type inference solution based on natural language for the prediction of these types.  ... 
arXiv:1912.00680v1 fatcat:4a4iqgp64fd3jmcj2vjgegz3oy

HiTyper: A Hybrid Static Type Inference Framework with Neural Prediction [article]

Yun Peng, Zongjie Li, Cuiyun Gao, Bowei Gao, David Lo, Michael Lyu
2021 arXiv   pre-print
To mitigate the challenges, in this paper, we propose a hybrid type inference framework named HiTyper, which integrates static inference into deep learning (DL) models for more accurate type prediction  ...  Type inference for dynamic programming languages is an important yet challenging task.  ...  For the slots that cannot be inferred by static analysis, HiTyper then adopts a deep learning (DL) model for further prediction.  ... 
arXiv:2105.03595v1 fatcat:x2u7yvdclrhdfmacipifewa3pi

Evaluation of Inference Attack Models for Deep Learning on Medical Data [article]

Maoqiang Wu, Xinyue Zhang, Jiahao Ding, Hien Nguyen, Rong Yu, Miao Pan, Stephen T. Wong
2020 arXiv   pre-print
Deep learning has attracted broad interest in healthcare and medical communities.  ...  The experimental evaluations show that our proposed defense approaches can effectively reduce the potential privacy leakage of medical deep learning from the inference attacks.  ...  Related Work There are different types of privacy attacks against training and inference data. These attacks severely threaten patients' privacy when deep learning is used in the healthcare area.  ... 
arXiv:2011.00177v1 fatcat:xjtspuby7jfjjardfmbd72siry

Full-Cycle Energy Consumption Benchmark for Low-Carbon Computer Vision [article]

Bo Li, Xinyang Jiang, Donglin Bai, Yuge Zhang, Ningxin Zheng, Xuanyi Dong, Lu Liu, Yuqing Yang, Dongsheng Li
2021 arXiv   pre-print
However, most of the existing efficient deep learning methods do not explicitly consider energy consumption as a key performance indicator.  ...  The benchmark can provide insights for low carbon emission when selecting efficient deep learning algorithms in different model usage scenarios.  ...  applications. 2 Related Work 2.1 Efficient Deep Learning Table 1 : 1 Life-cycle stages in different types of efficient deep learning methods.  ... 
arXiv:2108.13465v2 fatcat:gqiu7mvyhrawdipyhm6x3j77tq

Comparison of Type-2 Fuzzy Inference Method and Deep Neural Networks for Mass Detection from Breast Ultrasonography Images

2020 Cumhuriyet Science Journal  
The result of this study was compared with the result of another study that implemented type-2 fuzzy inference system with a success rate of 99,34%.  ...  As a conclusion, it can be expressed that the deep neural networks are more successful than fuzzy inference systems in tumour detection from breast ultrasonography images.  ...  : Figure 1 . 1 Type-2 fuzzy inference system. Figure 2 . 2 Deep neural network structure.  ... 
doi:10.17776/csj.691683 fatcat:vac75zjn3jhu5agmdfx33rh724

Process planning for die and mold machining based on pattern recognition and deep learning

2021 Journal of Advanced Mechanical Design, Systems, and Manufacturing  
learning.  ...  Therefore, this study aims to develop a CAPP system that can determine machining process information for complicated surfaces of die and mold based on pattern recognition and deep learning, a kind of machine  ...  Targeted shape Output data Tool path pattern Deep Learning Cutting tool type Input data Fig. 9 Inference flow of cutting tool type.  ... 
doi:10.1299/jamdsm.2021jamdsm0015 fatcat:mbfahmfzizgofphc2cfx67miea

Temporal Attribute Prediction via Joint Modeling of Multi-Relational Structure Evolution

Sankalp Garg, Navodita Sharma, Woojeong Jin, Xiang Ren
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Recently, there has been a focus on the application of deep representation learning on dynamic graphs.  ...  Time series prediction is an important problem in machine learning. Previous methods for time series prediction did not involve additional information.  ...  There are other recent works involving deep neural networks learning with inference.  ... 
doi:10.24963/ijcai.2020/382 dblp:conf/ijcai/GuoFZUK20 fatcat:tb6uey3tmnb2fo536rfbpdw77i

An Embedded System for Image-based Crack Detection by using Fine-Tuning model of Adaptive Structural Learning of Deep Belief Network [article]

Shin Kamada, Takumi Ichimura
2021 arXiv   pre-print
In our research, an adaptive structural learning method of Restricted Boltzmann Machine (Adaptive RBM) and Deep Belief Network (Adaptive DBN) have been developed as a deep learning model.  ...  Deep learning has been a successful model which can effectively represent several features of input space and remarkably improve image recognition performance on the deep architectures.  ...  The NVIDIA Jetson series are known to be a tiny embedded system which enables fast inference of deep learning.  ... 
arXiv:2110.13145v1 fatcat:k7ijosnnlrcjvk5g3icky4rn4i

An Early Resource Characterization of Deep Learning on Wearables, Smartphones and Internet-of-Things Devices

Nicholas D. Lane, Sourav Bhattacharya, Petko Georgiev, Claudio Forlivesi, Fahim Kawsar
2015 Proceedings of the 2015 International Workshop on Internet of Things towards Applications - IoT-App '15  
Deep Learning -is one of the most promising approaches for overcoming this challenge, and achieving more robust and reliable inference.  ...  Efforts to address this barrier to deep learning adoption are slowed by our lack of a systematic understanding of how these algorithms behave at inference time on resource constrained hardware.  ...  Enabling wide-spread device-side deep learning inference will require a range of brand-new techniques for optimized resource sensitive execution.  ... 
doi:10.1145/2820975.2820980 dblp:conf/sensys/LaneBGFK15 fatcat:m6omcl5synckzd7ukuvw77qaea

HG-Caffe: Mobile and Embedded Neural Network GPU (OpenCL) Inference Engine with FP16 Supporting [article]

Zhuoran Ji
2019 arXiv   pre-print
Breakthroughs in the fields of deep learning and mobile system-on-chips are radically changing the way we use our smartphones.  ...  In this paper, we present a deep neural network inference engine named HG-Caffe, which supports GPUs with half precision.  ...  Scalar Value All in Float In most deep learning frameworks, the data types for tensors (arrays) can be declared in float and double.  ... 
arXiv:1901.00858v1 fatcat:h66j7hes2vgmhapnb5uotazv5y
« Previous Showing results 1 — 15 out of 240,299 results