Recurrent Attentive Neural Process for Sequential Data

Qin, Shenghao; Zhu, Jiacheng; Qin, Jimmy; Wang, Wenshuo; Zhao, Ding

doi:10.48550/arxiv.1910.09323

Cited by 7 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Future work will explore improving the calibration of the downscaling models using mixture distributions and normalising flows (Rezende and Mohamed, 2015) to improve the calibration of the model. A further possibility for extending the convCNP model would be to explicitly incorporate time by building recurrence into the model (Qin et al, 2019;Singh et al, 2019).…”

Section: Discussionmentioning

confidence: 99%

Convolutional conditional neural processes for local climate downscaling

Vaughan¹,

Tebbutt²,

Hosking³

et al. 2021

Preprint

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

Section: Discussionmentioning

confidence: 99%

Convolutional conditional neural processes for local climate downscaling

Vaughan¹,

Tebbutt²,

Hosking³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…To define stochastic processes, we indeed need to ensure invariance to input permutations, i.e., exchangeability condition [19]. However, it is reported to be practically beneficial to relax such assumption when the observations contain time sequences [20]. Specifically, recurrent attentive neural processes (RANP) [20] incorporates a recurrent neural network structure to process the observations, and show improved performance on vehicle trajectory predicition over ANP.…”

Section: B Functional Distribution With Efficient Inferencementioning

confidence: 99%

“…However, it is reported to be practically beneficial to relax such assumption when the observations contain time sequences [20]. Specifically, recurrent attentive neural processes (RANP) [20] incorporates a recurrent neural network structure to process the observations, and show improved performance on vehicle trajectory predicition over ANP. We point out that in RANP, the exchangeability is only relaxed on observations, while the temporal structure of test-time input and output is not considered.…”

Section: B Functional Distribution With Efficient Inferencementioning

confidence: 99%

See 1 more Smart Citation

Learn from Human Teams: a Probabilistic Solution to Real-Time Collaborative Robot Handling with Dynamic Gesture Commands

Chen¹,

Alvin²,

Liu³

2021

Preprint

View full text Add to dashboard Cite

We study real-time collaborative robot (cobot) handling, where the cobot maneuvers a workpiece under human commands. This is useful when it is risky for humans to directly handle the workpiece. However, it is hard to make the cobot both easy to command and flexible in possible operations. In this work, we propose a Real-Time Collaborative Robot Handling (RTCoHand) framework that allows the control of cobot via user-customized dynamic gestures. This is hard due to variations among users, human motion uncertainties, and noisy human input. We model the task as a probabilistic generative process, referred to as Conditional Collaborative Handling Process (CCHP), and learn from human-human collaboration. We thoroughly evaluate the adaptability and robustness of CCHP and apply our approach to a real-time cobot handling task with Kinova Gen3 robot arm. We achieve seamless human-robot collaboration with both experienced and new users. Compared to classical controllers, RTCoHand allows significantly more complex maneuvers and lower user cognitive burden. It also eliminates the need for trialand-error, rendering it advantageous in safety-critical tasks.

show abstract

“…There have been proposed many methods, such as transfer learning, domain adaptation, multitask learning [52,33,20,18,25], and meta-learning [48,5,43,2,54,49,4,9,42,21,55,11], to improve performance using data from related but different domains for regression, classification, and time-series prediction [16,44,29,41,51,1,39,40,17]. However, these methods are inapplicable for Koopman spectral analysis.…”

Section: Related Workmentioning

confidence: 99%

Meta-Learning for Koopman Spectral Analysis with Short Time-series

Iwata¹,

Kawahara²

2021

Preprint

View full text Add to dashboard Cite

Koopman spectral analysis has attracted attention for nonlinear dynamical systems since we can analyze nonlinear dynamics with a linear regime by embedding data into a Koopman space by a nonlinear function. For the analysis, we need to find appropriate embedding functions. Although several neural network-based methods have been proposed for learning embedding functions, existing methods require long time-series for training neural networks. This limitation prohibits performing Koopman spectral analysis in applications where only short time-series are available.In this paper, we propose a meta-learning method for estimating embedding functions from unseen short time-series by exploiting knowledge learned from related but different time-series. With the proposed method, a representation of a given short time-series is obtained by a bidirectional LSTM for extracting its properties. The embedding function of the short time-series is modeled by a neural network that depends on the time-series representation. By sharing the LSTM and neural networks across multiple time-series, we can learn common knowledge from different timeseries while modeling time-series-specific embedding functions with the time-series representation.Our model is trained such that the expected test prediction error is minimized with the episodic training framework. We experimentally demonstrate that the proposed method achieves better performance in terms of eigenvalue estimation and future prediction than existing methods.

show abstract

Recurrent Attentive Neural Process for Sequential Data

Cited by 7 publications

References 15 publications

Convolutional conditional neural processes for local climate downscaling

Convolutional conditional neural processes for local climate downscaling

Learn from Human Teams: a Probabilistic Solution to Real-Time Collaborative Robot Handling with Dynamic Gesture Commands

Meta-Learning for Koopman Spectral Analysis with Short Time-series

Contact Info

Product

Resources

About