Toward the third generation artificial intelligence
There have been two competing paradigms in artificial intelligence (AI) development ever
since its birth in 1956, ie, symbolism and connectionism (or sub-symbolism). While …
since its birth in 1956, ie, symbolism and connectionism (or sub-symbolism). While …
Estimating high order gradients of the data distribution by denoising
The first order derivative of a data density can be estimated efficiently by denoising score
matching, and has become an important component in many applications, such as image …
matching, and has become an important component in many applications, such as image …
Efficient learning of generative models via finite-difference score matching
Several machine learning applications involve the optimization of higher-order derivatives
(eg, gradients of gradients) during training, which can be expensive with respect to memory …
(eg, gradients of gradients) during training, which can be expensive with respect to memory …
Functional variational inference based on stochastic process generators
C Ma, JM Hernández-Lobato - Advances in Neural …, 2021 - proceedings.neurips.cc
Bayesian inference in the space of functions has been an important topic for Bayesian
modeling in the past. In this paper, we propose a new solution to this problem called …
modeling in the past. In this paper, we propose a new solution to this problem called …
Algorithmic differentiation for automated modeling of machine learned force fields
Reconstructing force fields (FFs) from atomistic simulation data is a challenge since accurate
data can be highly expensive. Here, machine learning (ML) models can help to be data …
data can be highly expensive. Here, machine learning (ML) models can help to be data …
Nonparametric generative modeling with conditional sliced-Wasserstein flows
Sliced-Wasserstein Flow (SWF) is a promising approach to nonparametric generative
modeling but has not been widely adopted due to its suboptimal generative quality and lack …
modeling but has not been widely adopted due to its suboptimal generative quality and lack …
Filtering dynamical systems using observations of statistics
We consider the problem of filtering dynamical systems, possibly stochastic, using
observations of statistics. Thus, the computational task is to estimate a time-evolving density …
observations of statistics. Thus, the computational task is to estimate a time-evolving density …
MARS: Meta-learning as score matching in the function space
Meta-learning aims to extract useful inductive biases from a set of related datasets. In
Bayesian meta-learning, this is typically achieved by constructing a prior distribution over …
Bayesian meta-learning, this is typically achieved by constructing a prior distribution over …
[PDF][PDF] Approximate bayesian inference with stein functional variational gradient descent
We propose a general-purpose variational algorithm that forms a natural analogue of Stein
variational gradient descent (SVGD) in function space. While SVGD successively updates a …
variational gradient descent (SVGD) in function space. While SVGD successively updates a …