Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code that supports symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development tends to produce DL code that is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, less error-prone imperative DL frameworks encouraging eager execution have emerged but at the expense of run-time performance. While hybrid approaches aim for the "best of both worlds," the challenges in applying them in the real world are largely unknown. We conduct a data-driven analysis of challenges---and resultant bugs---involved in writing reliable yet performant imperative DL code by studying 250 open-source projects, consisting of 19.7 MLOC, along with 470 and 446 manually examined code patches and bug reports, respectively. The results indicate that hybridization: (i) is prone to API misuse, (ii) can result in performance degradation---the opposite of its intention, and (iii) has limited application due to execution mode incompatibility. We put forth several recommendations, best practices, and anti-patterns for effectively hybridizing imperative DL code, potentially benefiting DL practitioners, API designers, tool developers, and educators.
more »
« less
Challenges in Migrating Imperative Deep Learning Programs to Graph Execution: An Empirical Study
{"Abstract":["
Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code that supports symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development tends to produce DL code that is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, less error-prone imperative DL frameworks encouraging eager execution have emerged but at the expense of run-time performance. While hybrid approaches aim for the "best of both worlds," the challenges in applying them in the real world are largely unknown. We conduct a data-driven analysis of challenges\u2014and resultant bugs\u2014involved in writing reliable yet performant imperative DL code by studying 250 open-source projects, consisting of 19.7 MLOC, along with 470 and 446 manually examined code patches and bug reports, respectively. The results indicate that hybridization: (i) is prone to API misuse, (ii) can result in performance degradation\u2014the opposite of its intention, and (iii) has limited application due to execution mode incompatibility. We put forth several recommendations, best practices, and anti-patterns for effectively hybridizing imperative DL code, potentially benefiting DL practitioners, API designers, tool developers, and educators.<\/p>"],"Other":["Support for this project was provided by PSC-CUNY Award #638010051, jointly funded by The Professional Staff Congress and The City University of New York."]}
more » « less- Award ID(s):
- 2200343
- NSF-PAR ID:
- 10417286
- Publisher / Repository:
- Zenodo
- Date Published:
- Subject(s) / Keyword(s):
- Deep Learning Imperative Programming Hybrid Programming Paradigms Graph-based Execution Empirical Studies Software Evolution
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code-supporting symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, imperative DL frameworks encouraging eager execution have emerged at the expense of run-time performance. Though hybrid approaches aim for the “best of both worlds,” using them effectively requires subtle considerations to make code amenable to safe, accurate, and efficient graph execution. We present our ongoing work on automated refactoring that assists developers in specifying whether and how their otherwise eagerly-executed imperative DL code could be reliably and efficiently executed as graphs while preserving semantics. The approach, based on a novel imperative tensor analysis, will automatically determine when it is safe and potentially advantageous to migrate imperative DL code to graph execution and modify decorator parameters or eagerly executing code already running as graphs. The approach is being implemented as a PyDev Eclipse IDE plug-in and uses the WALA Ariadne analysis framework. We discuss our ongoing work towards optimizing imperative DL code to its full potential.more » « less
-
more » « less
Abstract An intelligent sensing framework using Machine Learning (ML) and Deep Learning (DL) architectures to precisely quantify dielectrophoretic force invoked on microparticles in a textile electrode-based DEP sensing device is reported. The prediction accuracy and generalization ability of the framework was validated using experimental results. Images of pearl chain alignment at varying input voltages were used to build deep regression models using modified ML and CNN architectures that can correlate pearl chain alignment patterns of Saccharomyces cerevisiae(yeast) cells and polystyrene microbeads to DEP force. Various ML models such as K-Nearest Neighbor, Support Vector Machine, Random Forest, Neural Networks, and Linear Regression along with DL models such as Convolutional Neural Network (CNN) architectures of AlexNet, ResNet-50, MobileNetV2, and GoogLeNet have been analyzed in order to build an effective regression framework to estimate the force induced on yeast cells and microbeads. The efficiencies of the models were evaluated using Mean Absolute Error, Mean Absolute Relative, Mean Squared Error, R-squared, and Root Mean Square Error (RMSE) as evaluation metrics. ResNet-50 with RMSPROP gave the best performance, with a validation RMSE of 0.0918 on yeast cells while AlexNet with ADAM optimizer gave the best performance, with a validation RMSE of 0.1745 on microbeads. This provides a baseline for further studies in the application of deep learning in DEP aided Lab-on-Chip devices.
-
null (Ed.)There is an increasing emphasis on securing deep learning (DL) inference pipelines for mobile and IoT applications with privacy-sensitive data. Prior works have shown that privacy-sensitive data can be secured throughout deep learning inferences on cloud-offloaded models through trusted execution environments such as Intel SGX. However, prior solutions do not address the fundamental challenges of securing the resource-intensive inference tasks on low-power, low-memory devices (e.g., mobile and IoT devices), while achieving high performance. To tackle these challenges, we propose SecDeep, a low-power DL inference framework demonstrating that both security and performance of deep learning inference on edge devices are well within our reach. Leveraging TEEs with limited resources, SecDeep guarantees full confidentiality for input and intermediate data, as well as the integrity of the deep learning model and framework. By enabling and securing neural accelerators, SecDeep is the first of its kind to provide trusted and performant DL model inferencing on IoT and mobile devices. We implement and validate SecDeep by interfacing the ARM NN DL framework with ARM TrustZone. Our evaluation shows that we can securely run inference tasks with 16× to 172× faster performance than no acceleration approaches by leveraging edge-available accelerators.more » « less
-
Deep learning (DL) algorithms have achieved significantly high performance in object detection tasks. At the same time, augmented reality (AR) techniques are transforming the ways that we work and connect with people. With the increasing popularity of online and hybrid learning, we propose a new framework for improving students’ learning experiences with electrical engineering lab equipment by incorporating the abovementioned technologies. The DL powered automatic object detection component integrated into the AR application is designed to recognize equipment such as multimeter, oscilloscope, wave generator, and power supply. A deep neural network model, namely MobileNet-SSD v2, is implemented for equipment detection using TensorFlow’s object detection API. When a piece of equipment is detected, the corresponding AR-based tutorial will be displayed on the screen. The mean average precision (mAP) of the developed equipment detection model is 81.4%, while the average recall of the model is 85.3%. Furthermore, to demonstrate practical application of the proposed framework, we develop a multimeter tutorial where virtual models are superimposed on real multimeters. The tutorial includes images and web links as well to help users learn more effectively. The Unity3D game engine is used as the primary development tool for this tutorial to integrate DL and AR frameworks and create immersive scenarios. The proposed framework can be a useful foundation for AR and machine-learning-based frameworks for industrial and educational training.more » « less
