An Empirical Study of Adder Neural Networks for Object Detection
6. prosince 2021
Řečníci
Xinghao Chen
Sprecher:in · 0 Follower:innen
Minjing Dong
Sprecher:in · 0 Follower:innen
Chunjing Xu
Sprecher:in · 0 Follower:innen
Yunhe Wang
Sprecher:in · 0 Follower:innen
O prezentaci
Adder neural networks (AdderNets) have shown impressive performance on image classification with only addition operations, which are more energy efficient than traditional convolutional neural networks built with multiplications. Compared with classification, there is a strong demand on reducing the energy consumption of modern object detectors via AdderNets for real-world applications such as autonomous driving and face detection. In this paper, we present an empirical study of AdderNets for object detection. We first reveal that the batch normalization statistics in the pre-trained adder backbone should not be frozen, since the relatively large feature variance of AdderNets. Moreover, we insert more shortcut connections in the neck part and design a new feature fusion architecture for avoiding the sparse features of adder layers. We present extensive ablation studies to explore several design choices of adder detectors. Comparisons with state-of-the-arts are conducted on COCO and PASCAL VOC benchmarks. Specifically, the proposed Adder FCOS achieves a 37.8Adder neural networks (AdderNets) have shown impressive performance on image classification with only addition operations, which are more energy efficient than traditional convolutional neural networks built with multiplications. Compared with classification, there is a strong demand on reducing the energy consumption of modern object detectors via AdderNets for real-world applications such as autonomous driving and face detection. In this paper, we present an empirical study of AdderNets for ob…
Organizátor
NeurIPS 2021
Konto · 1,9k Follower:innen
O organizátorovi (NeurIPS 2021)
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
ScaleCert: Scalable Certified Defense against Adversarial Patches with Sparse Superficial Layers
Später ansehen
Husheng Han, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Fair Classification with Adversarial Perturbations
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
What has Changed Over the Last 5 Years That Makes This an Exciting Time?
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Damped Anderson Mixing for Deep Reinforcement Learning: Acceleration, Convergence, and Stabilization
Später ansehen
Ke Sun, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%
On the interplay between data structure and loss function in classification problems
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Deep Reinforcement Learning
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%