Convolutional Conditional Neural Processes

Gordon, Jonathan; Bruinsma, Wessel P.; Foong, Andrew Y. K.; Requeima, James; Dubois, Yann; Turner, Richard E.

doi:10.48550/arxiv.1910.13556

Cited by 8 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A later study [32] proposed Attentive Neural Processes (ANPs) by incorporating attention into NPs to alleviate the underfitting problem and improve the regression performance. NP-based models have shown great performance in function regression [33], image reconstruction [32], and point-cloud modeling [34]. As probabilistic latent variable models, ANPs naturally enable continual online learning in continuously parameterized environments.…”

Section: Related Workmentioning

confidence: 99%

Context-Aware Safe Reinforcement Learning for Non-Stationary Environments

Chen¹,

Liu²,

Zhu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Safety is a critical concern when deploying reinforcement learning agents for realistic tasks. Recently, safe reinforcement learning algorithms have been developed to optimize the agent's performance while avoiding violations of safety constraints. However, few studies have addressed the non-stationary disturbances in the environments, which may cause catastrophic outcomes. In this paper, we propose the context-aware safe reinforcement learning (CASRL) method, a meta-learning framework to realize safe adaptation in nonstationary environments. We use a probabilistic latent variable model to achieve fast inference of the posterior environment transition distribution given the context data. Safety constraints are then evaluated with uncertainty-aware trajectory sampling. The high cost of safety violations leads to the rareness of unsafe records in the dataset. We address this issue by enabling prioritized sampling during model training and formulating prior safety constraints with domain knowledge during constrained planning. The algorithm is evaluated in realistic safety-critical environments with non-stationary disturbances. Results show that the proposed algorithm significantly outperforms existing baselines in terms of safety and robustness.

show abstract

Section: Related Workmentioning

confidence: 99%

Context-Aware Safe Reinforcement Learning for Non-Stationary Environments

Chen¹,

Liu²,

Zhu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…For both variables the model produces qualitatively realistic time-series. This model is an example of a more general class of models known as convolutional conditional neural processes (ConvC-NPs; Gordon et al, 2019). Formally, PP downscaling is an instance of a supervised learning problem to learn the function f in equation 1.…”

Section: Inclusion Of Sub-grid Scale Topographymentioning

confidence: 99%

“…In the context of spatial data such as downscaling predictions, a desirable characteristic of predictions are that they are translationequivariant, that is the model makes identical predictions if data are spatially translated. The convCNP model (Gordon et al, 2019) applied here builds this equivariance into the CNP model. Throughout this study, we refer to the model developed in this section as 'the convCNP model'.…”

Section: Inclusion Of Sub-grid Scale Topographymentioning

confidence: 99%

Convolutional conditional neural processes for local climate downscaling

Vaughan¹,

Tebbutt²,

Hosking³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

A new model is presented for multisite statistical downscaling of temperature and precipitation using convolutional conditional neural processes (convCNPs). ConvCNPs are a recently developed class of models that allow deep learning techniques to be applied to off-the-grid spatio-temporal data. This model has a substantial advantage over existing downscaling methods in that the trained model can be used to generate multisite predictions at an arbitrary set of locations, regardless of the availability of training data. The convCNP model is shown to outperform an ensemble of existing downscaling techniques over Europe for both temperature and precipitation taken from the VALUE intercomparison project. The model also outperforms an approach that uses Gaussian processes to interpolate single-site downscaling models at unseen locations. Importantly, substantial improvement is seen in the representation of extreme precipitation events. These results indicate that the convCNP is a robust downscaling model suitable for generating localised projections for use in climate impact studies, and motivates further research into applications of deep learning techniques in statistical downscaling.

show abstract

“…Symmetry is mathematically described in terms of groups and has become an essential concept in machine learning. Gordon et al (2019) point out that, when data symmetry is represented by a infinite group like the translation group, equivariant maps, which are symmetry-preserving processing, cannot be captured as maps between finite-dimensional spaces but can be described by maps between infinite-dimensional function spaces. As a related study about symmetry-preserving processing, Finzi et al (2020) propose group convolution of functional representations and investigate practical computational methods such as discretization and localization.…”

Section: Related Workmentioning

confidence: 99%

Universal Approximation Theorem for Equivariant Maps by Group CNNs

Kumagai¹,

Sannai²

2020

Preprint

View full text Add to dashboard Cite

Group symmetry is inherent in a wide variety of data distributions. Data processing that preserves symmetry is described as an equivariant map and often effective in achieving high performance. Convolutional neural networks (CNNs) have been known as models with equivariance and shown to approximate equivariant maps for some specific groups. However, universal approximation theorems for CNNs have been separately derived with individual techniques according to each group and setting. This paper provides a unified method to obtain universal approximation theorems for equivariant maps by CNNs in various settings. As its significant advantage, we can handle non-linear equivariant maps between infinite-dimensional spaces for non-compact groups.

show abstract

Convolutional Conditional Neural Processes

Cited by 8 publications

References 15 publications

Context-Aware Safe Reinforcement Learning for Non-Stationary Environments

Context-Aware Safe Reinforcement Learning for Non-Stationary Environments

Convolutional conditional neural processes for local climate downscaling

Universal Approximation Theorem for Equivariant Maps by Group CNNs

Contact Info

Product

Resources

About