nGraph-HE

Boemer, Fabian; Lao, Yixing; Cammarota, Rosario; Wierzynski, Casimir

doi:10.1145/3310273.3323047

Cited by 111 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is limited to the interaction between the processor and memory leaving aside accelerator devices such as the GPU. In order to solve or significantly reduce this problem, methods have emerged such as the optimization of the directed asynchronous graph created at the time of the execution of the model, as proposed by Le [15] and Boemer [3] or, methods of working with sparses matrices [20] to reduce memory consumption during training processes. They have also considered changing the way, memory levels are used during training as mentioned in Rhu [21] and Lim [17] works but they also present bottlenecks due to the high level of communication that occurs through the PCIe bus.…”

Section: Discussionmentioning

confidence: 99%

“…All of the above have forced accelerator designers for deep neural networks to use highcost memory solutions such as HBM (High Bandwidth Memory) used in Google TPUs [11]. Other solutions have been proposed to overcome these limitations such as the development of new techniques to improve training by working directly on the neural network graph [3,15] or working with sparse matrices [20]. Designing specialized dense nodes for the effective use of accelerators has also been done.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Computational Resource Consumption in Convolutional Neural Network Training – A Focus on Memory

Barrio-Parra

Barrios

Denneulin

2021

JSFI

View full text Add to dashboard Cite

Deep neural networks (DNNs) have grown in popularity in recent years thanks to the increase in computing power and the size and relevance of data sets. This has made it possible to build more complex models and include more areas of research and application. At the same time, the amount of data generated during the training process of these models puts great pressure on the capacity and bandwidth of the memory subsystem and, as a direct consequence, has become one of the biggest bottlenecks for the scalability of neural networks. Therefore, the optimizing of the workloads produced by DNNs in the memory subsystem requires a detailed understanding of access to the memory and the interactions between the processor, accelerator devices, and the system memory hierarchy. However, contrary to what would be expected, most DNN profilers work at a high level, so they only perform an analysis of the model and individual layers of the network leaving aside the complex interactions between all the hardware components involved in the training. This article shows the characterization performed using a convolutional neural network implemented in the two most popular frameworks: TensorFlow and Pytorch. Likewise, the behavior of the component interactions is discussed by varying the batch size for two sets of synthetic data and showing the results obtained by the profiler created for the study. Moreover, the results obtained when evaluating the AlexNet version on TensorFlow and its similarity in behavior when using a basic CNN are included.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Computational Resource Consumption in Convolutional Neural Network Training – A Focus on Memory

Barrio-Parra

Barrios

Denneulin

2021

JSFI

View full text Add to dashboard Cite

show abstract

“…The nGraph-HE framework, proposed by Boemer et al [75] in 2019, is based on Intel's nGraph ML compiler [76] and translates standard TensorFlow computations into arithmetic circuits in BFV or CKKS using the SEAL library. It enables inference on pre-trained models over encrypted inputs, applying FHE-specific optimizations (e.g., constant folding, SIMD-packing, and graph-level optimizations such as lazy rescaling and depth-aware encoding), and run-time optimizations (e.g., bypassing special plaintext values).…”

Section: G Ngraph-hementioning

confidence: 99%

SoK: Fully Homomorphic Encryption Compilers

Viand

Jattke

Hithnawi

2021

2021 IEEE Symposium on Security and Privacy (SP)

View full text Add to dashboard Cite

Fully Homomorphic Encryption (FHE) allows a third party to perform arbitrary computations on encrypted data, learning neither the inputs nor the computation results. Hence, it provides resilience in situations where computations are carried out by an untrusted or potentially compromised party. This powerful concept was first conceived by Rivest et al. in the 1970s. However, it remained unrealized until Craig Gentry presented the first feasible FHE scheme in 2009.The advent of the massive collection of sensitive data in cloud services, coupled with a plague of data breaches, moved highly regulated businesses to increasingly demand confidential and secure computing solutions. This demand, in turn, has led to a recent surge in the development of FHE tools. To understand the landscape of recent FHE tool developments, we conduct an extensive survey and experimental evaluation to explore the current state of the art and identify areas for future development.In this paper, we survey, evaluate, and systematize FHE tools and compilers. We perform experiments to evaluate these tools' performance and usability aspects on a variety of applications. We conclude with recommendations for developers intending to develop FHE-based applications and a discussion on future directions for FHE tools development.

show abstract

“…This technique can directly adapt computational operations to encrypted data. A large number of methods have been proposed that automatically classify data encrypted with secure computation [ 4 , 5 , 6 ]. These methods can protect test data; however, the encrypted data can hardly be compressed.…”

Section: Introductionmentioning

confidence: 99%

Effects of JPEG Compression on Vision Transformer Image Classification for Encryption-then-Compression Images

Hamano

Imaizumi

Kiya

2023

Sensors

View full text Add to dashboard Cite

This paper evaluates the effects of JPEG compression on image classification using the Vision Transformer (ViT). In recent years, many studies have been carried out to classify images in the encrypted domain for privacy preservation. Previously, the authors proposed an image classification method that encrypts both a trained ViT model and test images. Here, an encryption-then-compression system was employed to encrypt the test images, and the ViT model was preliminarily trained by plain images. The classification accuracy in the previous method was exactly equal to that without any encryption for the trained ViT model and test images. However, even though the encrypted test images can be compressible, the practical effects of JPEG, which is a typical lossy compression method, have not been investigated so far. In this paper, we extend our previous method by compressing the encrypted test images with JPEG and verify the classification accuracy for the compressed encrypted-images. Through our experiments, we confirm that the amount of data in the encrypted images can be significantly reduced by JPEG compression, while the classification accuracy of the compressed encrypted-images is highly preserved. For example, when the quality factor is set to 85, this paper shows that the classification accuracy can be maintained at over 98% with a more than 90% reduction in the amount of image data. Additionally, the effectiveness of JPEG compression is demonstrated through comparison with linear quantization. To the best of our knowledge, this is the first study to classify JPEG-compressed encrypted images without sacrificing high accuracy. Through our study, we have come to the conclusion that we can classify compressed encrypted-images without degradation to accuracy.

show abstract

nGraph-HE

Cited by 111 publications

References 15 publications

Computational Resource Consumption in Convolutional Neural Network Training – A Focus on Memory

Computational Resource Consumption in Convolutional Neural Network Training – A Focus on Memory

SoK: Fully Homomorphic Encryption Compilers

Effects of JPEG Compression on Vision Transformer Image Classification for Encryption-then-Compression Images

Contact Info

Product

Resources

About