Patrick Kidger
8 papers · 2019–2021 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
π£ Hot Topic Early Bird π Conference Polyglot (4) π Cross-Pollinator (11) πΊοΈ Taxonomy Completionist (14) π Interdisciplinary Bridge
π§
Keyword Pioneer
π
Triple Crown
π
Keyword Champion
(2)
β‘
Prolific Year
(5)
Conferences
ICML (3)
NIPS (3)
COLT (1)
ICLR (1)
Top co-authors
Keywords
neural network
(3)
time series
(3)
neural differential equation
(3)
adjoint method
(2)
brownian motion
(2)
stochastic differential equation
(2)
controlled differential equation
(2)
generative adversarial network
(1)
deep learning
(1)
function approximation
(1)
variational autoencoder
(1)
recurrent neural network
(1)
data stream
(1)
adaptive step size
(1)
deep learning theory
(1)
numerical solver
(1)
continuous time
(1)
rough path theory
(1)
activation function
(1)
multivariate time series
(1)
Papers
Neural Rough Differential Equations for Long Time Series
ICML 2021
Signatory: differentiable computations of the signature and logsignature transforms, on both CPU and GPU
ICLR 2021
"Hey, thatβs not an ODE": Faster ODE Adjoints via Seminorms
ICML 2021
Neural SDEs as Infinite-Dimensional GANs
ICML 2021
Efficient and Accurate Gradients for Neural SDEs
NIPS 2021
Neural Controlled Differential Equations for Irregular Time Series
NIPS 2020
Universal Approximation with Deep Narrow Networks
COLT 2020
Deep Signature Transforms
NIPS 2019