Zobrazeno 1 - 10
of 18
pro vyhledávání: '"Ainsworth, Samuel"'
The success of deep learning is due in large part to our ability to solve certain massive non-convex optimization problems with relative ease. Though non-convex optimization is NP-hard, simple algorithms -- often variants of stochastic gradient desce
Externí odkaz:
http://arxiv.org/abs/2209.04836
Publikováno v:
L4DC 2021
We study the estimation of policy gradients for continuous-time systems with known dynamics. By reframing policy learning in continuous-time, we show that it is possible construct a more efficient and accurate gradient estimator. The standard back-pr
Externí odkaz:
http://arxiv.org/abs/2012.06684
Data scientists have relied on samples to analyze populations of interest for decades. Recently, with the increase in the number of public data repositories, sample data has become easier to access. It has not, however, become easier to analyze. This
Externí odkaz:
http://arxiv.org/abs/1912.07777
Publikováno v:
NeurIPS 2019
In many environments, only a relatively small subset of the complete state space is necessary in order to accomplish a given task. We develop a simple technique using emergency stops (e-stops) to exploit this phenomenon. Using e-stops significantly i
Externí odkaz:
http://arxiv.org/abs/1912.01649
Many problems in machine learning and related application areas are fundamentally variants of conditional modeling and sampling across multi-aspect data, either multi-view, multi-modal, or simply multi-group. For example, sampling from the distributi
Externí odkaz:
http://arxiv.org/abs/1806.09060
Deep generative models have recently yielded encouraging results in producing subjectively realistic samples of complex data. Far less attention has been paid to making these generative models interpretable. In many scenarios, ranging from scientific
Externí odkaz:
http://arxiv.org/abs/1802.06765
Real-world engineering systems are typically compared and contrasted using multiple metrics. For practical machine learning systems, performance tuning is often more nuanced than minimizing a single expected loss objective, and it may be more realist
Externí odkaz:
http://arxiv.org/abs/1612.04453
Publikováno v:
Infection and Immunity; August 1999, Vol. 67 Issue: 8 p3793-3799, 7p
Publikováno v:
In Clinical Microbiology Newsletter 1994 16(8):61-62