Deep Convolutional Gaussian Processes

被引：16

作者：

Blomqvist, Kenneth ^{[1
,2
]}

Kaski, Samuel ^{[1
,2
]}

Heinonen, Markus ^{[1
,2
]}

机构：

[1] Aalto Univ, Espoo, Finland

[2] Helsinki Inst Informat Technol HIIT, Espoo, Finland

来源：

MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2019, PT II | 2020年 / 11907卷

关键词：

Gaussian processes; Convolutions; Variational inference;

D O I：

10.1007/978-3-030-46147-8_35

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We propose deep convolutional Gaussian processes, a deep Gaussian process architecture with convolutional structure. The model is a principled Bayesian framework for detecting hierarchical combinations of local features for image classification. We demonstrate greatly improved image classification performance compared to current convolutional Gaussian process approaches on the MNIST and CIFAR-10 datasets. In particular, we improve state-of-the-art CIFAR-10 accuracy by over 10% points.

引用

页码：582 / 597

页数：16

共 50 条

[31] Random Feature Expansions for Deep Gaussian Processes
Cutajar, Kurt
Bonilla, Edwin, V
Michiardi, Pietro
Filippone, Maurizio
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 70, 2017, 70
[32] Deep state-space Gaussian processes
Zhao, Zheng
Emzir, Muhammad
Sarkka, Simo
STATISTICS AND COMPUTING, 2021, 31 (06)
[33] Stochastic Deep Gaussian Processes over Graphs
Li, Naiqi
Li, Wenjie
Sun, Jifeng
Gao, Yinghua
Jiang, Yong
Xia, Shu-Tao
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
[34] Bayesian Uncertainty Estimation in Landmark Localization Using Convolutional Gaussian Processes
Schobs, Lawrence
McDonald, Thomas M.
Lu, Haiping
UNCERTAINTY FOR SAFE UTILIZATION OF MACHINE LEARNING IN MEDICAL IMAGING, UNSURE 2023, 2023, 14291 : 22 - 31
[35] On the use of convolutional Gaussian processes to improve the seasonal forecasting of precipitation and temperature
Wang, Chao
Zhang, Wei
Villarini, Gabriele
JOURNAL OF HYDROLOGY, 2021, 593 (593)
[36] Forecasting global climate drivers using Gaussian processes and convolutional autoencoders
Donnelly, James
Daneshkhah, Alireza
Abolfathi, Soroush
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 128
[37] Deep Gaussian Processes for Calibration of Computer Models (with Discussion)*
Marmin, Sebastien
Filippone, Maurizio
BAYESIAN ANALYSIS, 2022, 17 (04): : 1301 - 1330
[38] Wide neural networks with bottlenecks are deep gaussian processes
Agrawal, Devanshu
Papamarkou, Theodore
Hinkle, Jacob
Journal of Machine Learning Research, 2020, 21
[39] LEARNING FROM HETEROGENEOUS DATA WITH DEEP GAUSSIAN PROCESSES
Ajirak, Marzieh
Preis, Heidi
Lobel, Marci
Djuric, Petar M.
2023 IEEE 9TH INTERNATIONAL WORKSHOP ON COMPUTATIONAL ADVANCES IN MULTI-SENSOR ADAPTIVE PROCESSING, CAMSAP, 2023, : 46 - 50
[40] Approximate Inference Turns Deep Networks into Gaussian Processes
Khan, Mohammad Emtiyaz
Immer, Alexander
Abedi, Ehsan
Korzepa, Maciej
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 32

← 1 2 3 4 5 →