Zobrazeno 1 - 10
of 55
pro vyhledávání: '"Mnih, Andriy"'
Neural Posterior Estimation methods for simulation-based inference can be ill-suited for dealing with posterior distributions obtained by conditioning on multiple observations, as they tend to require a large number of simulator calls to learn accura
Externí odkaz:
http://arxiv.org/abs/2209.14249
Training large-scale mixture of experts models efficiently on modern hardware requires assigning datapoints in a batch to different experts, each with a limited capacity. Recently proposed assignment procedures lack a probabilistic interpretation and
Externí odkaz:
http://arxiv.org/abs/2109.11817
Training models with discrete latent variables is challenging due to the high variance of unbiased gradient estimators. While low-variance reparameterization gradients of a continuous relaxation can provide an effective solution, a continuous relaxat
Externí odkaz:
http://arxiv.org/abs/2106.08056
Autor:
Bauer, Matthias, Mnih, Andriy
Publikováno v:
38th International Conference on Machine Learning (ICML 2021)
Efficient low-variance gradient estimation enabled by the reparameterization trick (RT) has been essential to the success of variational autoencoders. Doubly-reparameterized gradients (DReGs) improve on the RT for multi-sample variational bounds by a
Externí odkaz:
http://arxiv.org/abs/2101.11046
Publikováno v:
Part of Advances in Neural Information Processing Systems 33 proceedings (NeurIPS 2020)
Training models with discrete latent variables is challenging due to the difficulty of estimating the gradients accurately. Much of the recent progress has been achieved by taking advantage of continuous relaxations of the system, which are not alway
Externí odkaz:
http://arxiv.org/abs/2006.10680
Lipschitz constants of neural networks have been explored in various contexts in deep learning, such as provable adversarial robustness, estimating Wasserstein distance, stabilising training of GANs, and formulating invertible neural networks. Such w
Externí odkaz:
http://arxiv.org/abs/2006.04710
Applying Q-learning to high-dimensional or continuous action spaces can be difficult due to the required maximization over the set of possible actions. Motivated by techniques from amortized inference, we replace the expensive maximization over all a
Externí odkaz:
http://arxiv.org/abs/2001.08116
We introduce a new interpretation of sparse variational approximations for Gaussian processes using inducing points, which can lead to more scalable algorithms than previous methods. It is based on decomposing a Gaussian process as a sum of two indep
Externí odkaz:
http://arxiv.org/abs/1910.10596
Publikováno v:
Journal of Machine Learning Research, 21(132):1-62, 2020
This paper is a broad and accessible survey of the methods we have at our disposal for Monte Carlo gradient estimation in machine learning and across the statistical sciences: the problem of computing the gradient of an expectation of a function with
Externí odkaz:
http://arxiv.org/abs/1906.10652
Autor:
Kim, Hyunjik, Mnih, Andriy, Schwarz, Jonathan, Garnelo, Marta, Eslami, Ali, Rosenbaum, Dan, Vinyals, Oriol, Teh, Yee Whye
Neural Processes (NPs) (Garnelo et al 2018a;b) approach regression by learning to map a context set of observed input-output pairs to a distribution over regression functions. Each function models the distribution of the output given an input, condit
Externí odkaz:
http://arxiv.org/abs/1901.05761