Deep Convolutional Neural Networks

Ju, Liu Bie

doi:10.1002/047134608x.w8424

“…First, the initial convolutional layer is now followed by spatial dropout [91] (with p = 0.1). Second, we attach the ECA gate [69] (with k = 3) before the Localization Classification Regression (LCR) layer.…”

Section: Methodsmentioning

confidence: 99%

“…On the contrary, most of the attention modules inevitably increase model complexity. Efficient Channel Attention (ECA) gate [69] is a soft attention mechanism that addresses this issue. It avoids dimensionality reduction and captures cross-channel interaction efficiently.…”

Section: Attention Mechanismsmentioning

confidence: 99%

Lightweight Jet Reconstruction and Identification as an Object Detection Task

Pol

¹

,

Aarrestad

²

,

Говоркова

³

et al. 2022

Preprint

0

View full text Add to dashboard Cite

We apply object detection techniques based on deep convolutional blocks to end-to-end jet identification and reconstruction tasks encountered at the CERN Large Hadron Collider (LHC). Collision events produced at the LHC and represented as an image composed of calorimeter and tracker cells are given as an input to a Single Shot Detection network. The algorithm, named PFJet-SSD performs simultaneous localization, classification and regression tasks to cluster jets and reconstruct their features. This all-in-one single feed-forward pass gives advantages in terms of execution time and an improved accuracy w.r.t. traditional rule-based methods. A further gain is obtained from network slimming, homogeneous quantization, and optimized runtime for meeting memory and latency constraints of a typical real-time processing environment. We experiment with 8-bit and ternary quantization, benchmarking their accuracy and inference latency against a single-precision floating-point. We show that the ternary network closely matches the performance of its full-precision equivalent and outperforms the state-of-the-art rule-based algorithm. Finally, we report the inference latency on different hardware platforms and discuss future applications.

show abstract

“…First, the initial convolutional layer is now followed by spatial dropout [94] (with p = 0.1). Second, we attach the ECA gate [72] (with k = 3) before the localization classification regression (LCR) layer.…”

Section: Methodsmentioning

confidence: 99%

“…On the contrary, most of the attention modules inevitably increase model complexity. Efficient channel attention (ECA) gate [72] is a soft attention mechanism that addresses this issue. It avoids dimensionality reduction and captures cross-channel interaction efficiently.…”

Section: Attention Mechanismsmentioning

confidence: 99%

Lightweight jet reconstruction and identification as an object detection task

Pol¹,

Aarrestad²,

Govorkova³

et al. 2022

Mach. Learn.: Sci. Technol.

2

0

View full text Add to dashboard Cite

We apply object detection techniques based on deep convolutional blocks to end-to-end jet identification and reconstruction tasks encountered at the CERN Large Hadron Collider (LHC). Collision events produced at the LHC and represented as an image composed of calorimeter and tracker cells are given as an input to a Single Shot Detection network. The algorithm, named PFJet-SSD performs simultaneous localization, classification and regression tasks to cluster jets and reconstruct their features. This all-in-one single feed-forward pass gives advantages in terms of execution time and an improved accuracy w.r.t. traditional rule-based methods. A further gain is obtained from network slimming, homogeneous quantization, and optimized runtime for meeting memory and latency constraints of a typical real-time processing environment. We experiment with 8-bit and ternary quantization, benchmarking their accuracy and inference latency against a single-precision floating-point. We show that the ternary network closely matches the performance of its full-precision equivalent and outperforms the state-of-the-art rule-based algorithm. Finally, we report the inference latency on different hardware platforms and discuss future applications.

show abstract

Approximation of functions from Korobov spaces by deep convolutional neural networks

Mao

¹

2022

View full text Add to dashboard Cite

The efficiency of deep convolutional neural networks (DCNNs) has been demonstrated empirically in many practical applications. In this paper, we establish a theory for approximating functions from Korobov spaces by DCNNs. It verifies rigorously the efficiency of DCNNs in approximating functions of many variables with some variable structures and their abilities in overcoming the curse of dimensionality.

show abstract

Deep Convolutional Neural Networks

Cited by 5 publications

References 35 publications

Lightweight Jet Reconstruction and Identification as an Object Detection Task

Lightweight Jet Reconstruction and Identification as an Object Detection Task

Lightweight jet reconstruction and identification as an object detection task

Approximation of functions from Korobov spaces by deep convolutional neural networks

Contact Info

Product

Resources

About