Match!

Aging Evolution for Image Classifier Architecture Search

Published on Jan 27, 2019 in AAAI (National Conference on Artificial Intelligence)
Esteban Real10
Estimated H-index: 10
(Google),
Alok Aggarwal3
Estimated H-index: 3
(Google)
+ 1 AuthorsQuoc V. Le58
Estimated H-index: 58
(Google)
Sources
Abstract
  • References (0)
  • Citations (8)
📖 Papers frequently viewed together
2018CVPR: Computer Vision and Pattern Recognition
4 Authors (Barret Zoph, ..., Quoc V. Le)
712 Citations
2018ICML: International Conference on Machine Learning
5 Authors (Gabriel Bender, ..., Quoc V. Le)
62 Citations
2017ICML: International Conference on Machine Learning
8 Authors (Esteban Real, ..., Alexey Kurakin)
108 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
References0
Newest
Cited By8
Newest
Apr 30, 2020 in ICLR (International Conference on Learning Representations)
#1Jonathan S. Rosenfeld (MIT: Massachusetts Institute of Technology)H-Index: 1
#2Amir Rosenfeld (MIT: Massachusetts Institute of Technology)H-Index: 5
Last. Nir Shavit (York University)H-Index: 49
view all 4 authors...
The dependency of the generalization error of neural networks on model and dataset size is of critical importance both in practice and for understanding the theory of neural networks. Nevertheless, the functional form of this dependency remains elusive. In this work, we present a functional form which approximates well the generalization error in practice. Capitalizing on the successful concept of model scaling (e.g., width, depth), we are able to simultaneously construct such a form and specify...
4 Citations
Apr 30, 2020 in ICLR (International Conference on Learning Representations)
#1Arber ZelaH-Index: 2
#2Thomas Elsken (Bosch)H-Index: 7
Last. Frank Hutter (University of Freiburg)H-Index: 40
view all 6 authors...
Differentiable Architecture Search (DARTS) has attracted a lot of attention due to its simplicity and small search costs achieved by a continuous relaxation and an approximation of the resulting bi-level optimization problem. However, DARTS does not work robustly for new problems: we identify a wide range of search spaces for which DARTS yields degenerate architectures with very poor test performance. We study this failure mode and show that, while DARTS successfully minimizes validation loss, t...
8 Citations
Apr 30, 2020 in ICLR (International Conference on Learning Representations)
#1Arber ZelaH-Index: 2
#2Julien SiemsH-Index: 1
Last. Frank Hutter (University of Freiburg)H-Index: 40
view all 3 authors...
One-shot neural architecture search (NAS) has played a crucial role in making NAS methods computationally feasible in practice. Nevertheless, there is still a lack of understanding on how these weight-sharing algorithms exactly work due to the many factors controlling the dynamics of the process. In order to allow a scientific study of these components, we introduce a general framework for one-shot NAS that can be instantiated to many recently-introduced variants and introduce a general benchmar...
7 Citations
#1Wenshuo Li (THU: Tsinghua University)H-Index: 3
#1Wenshuo Li (THU: Tsinghua University)H-Index: 1
Last. Huazhong Yang (THU: Tsinghua University)H-Index: 29
view all 6 authors...
With the fast evolvement of deep-learning specific embedded computing systems, applications powered by deep learning are moving from the cloud to the edge. When deploying NNs onto the edge devices under complex environments, there are various types of possible faults: soft errors caused by atmospheric neutrons and radioactive impurities, voltage instability, aging, temperature variations, and malicious attackers. Thus the safety risk of deploying neural networks at edge computing devices in safe...
Source
#1Xuefei NingH-Index: 3
#2Guangjun GeH-Index: 1
Last. Huazhong YangH-Index: 29
view all 9 authors...
With the fast evolvement of embedded deep-learning computing systems, applications powered by deep learning are moving from the cloud to the edge. When deploying neural networks (NNs) onto the devices under complex environments, there are various types of possible faults: soft errors caused by cosmic radiation and radioactive impurities, voltage instability, aging, temperature variations, and malicious attackers. Thus the safety risk of deploying NNs is now drawing much attention. In this paper,...
Most of the recent advances in crowd counting have evolved from hand-designed density estimation networks, where multi-scale features are leveraged to address scale variation, but at the expense of demanding design efforts. In this work, we automate the design of counting models with Neural Architecture Search (NAS) and introduce an end-to-end searched encoder-decoder architecture, Automatic Multi-Scale Network (AMSNet). The encoder and decoder in AMSNet are composed of different cells discovere...
We present a neural architecture search (NAS) technique to enhance the performance of unsupervised image de-noising, in-painting and super-resolution under the recently proposed Deep Image Prior (DIP). We show that evolutionary search can automatically optimize the encoder-decoder (E-D) structure and meta-parameters of the DIP network, which serves as a content-specific prior to regularize these single image restoration tasks. Our binary representation encodes the design space for an asymmetric ...
1 Citations
Multi-task learning is an open and challenging problem in computer vision. The typical way of conducting multi-task learning with deep neural networks is either through handcrafting schemes that share all initial layers and branch out at an adhoc point or through using separate task-specific networks with an additional feature sharing/fusion mechanism. Unlike existing methods, we propose an adaptive sharing approach, called AdaShare, that decides what to share across which tasks for achieving th...
#1Thomas ElskenH-Index: 7
#2Benedikt StafflerH-Index: 1
Last. Frank HutterH-Index: 40
view all 4 authors...
The recent progress in neural architectures search (NAS) has allowed scaling the automated design of neural architectures to real-world domains such as object detection and semantic segmentation. However, one prerequisite for the application of NAS are large amounts of labeled data and compute resources. This renders its application challenging in few-shot learning scenarios, where many related tasks need to be learned, each with limited amounts of data and compute time. Thus, few-shot learning ...
2 Citations
#1Xue Gu (JLU: Jilin University)
#2Ziyao Meng (JLU: Jilin University)
Last. Chunguo Wu (JLU: Jilin University)H-Index: 6
view all 7 authors...
Although deep neural networks (DNNs) play important roles in many fields, the architecture design of DNNs can be challenging due to the difficulty of input data representation, the huge number of parameters and the complex layer relationships. To overcome the obstacles of architecture design, we developed a new method to generate the optimal structure of DNNs, named Evolutionary Strategy-based Architecture Evolution (ESAE), consisting of a bi-level representation and a probability distribution l...
Source