Home /
Research
Showing 61 - 66 / 897
The Advantage of Conditional Meta-Learning for Biased Regularization and Fine Tuning
GiuliaDenevi, MassimilianoPontil, CarloCiliberto....
Published date-12/01/2020
Meta-Learning
Biased regularization and fine tuning are two recent meta-learning approaches. They have been shown to be effective to tackle distributions of tasks, in which the tasks’ target vectors are all …
Disentangling Label Distribution for Long-tailed Visual Recognition
YoungkyuHong, SeungjuHan, KwangheeChoi....
Published date-12/01/2020
The current evaluation protocol of long-tailed visual recognition trains the classification model on the long-tailed source label distribution and evaluates its performance on the uniform target label distribution. Such protocol …
Optimal Variance Control of the Score-Function Gradient Estimator for Importance-Weighted Bounds
ValentinLiévin, AndreaDittadi, AndersChristensen....
Published date-12/01/2020
This paper introduces novel results for the score-function gradient estimator of the importance-weighted variational bound (IWAE). We prove that in the limit of large $K$ (number of importance samples) one …
Almost Surely Stable Deep Dynamics
NathanLawrence, PhilipLoewen, MichaelForbes....
Published date-12/01/2020
We introduce a method for learning provably stable deep neural network based dynamic models from observed data. Specifically, we consider discrete-time stochastic dynamic models, as they are of particular interest …
Improving model calibration with accuracy versus uncertainty optimization
RanganathKrishnan, OmeshTickoo....
Published date-12/01/2020
ImageClassification, VariationalInference
Obtaining reliable and accurate quantification of uncertainty estimates from deep neural networks is important in safety-critical applications. A well-calibrated model should be accurate when it is certain about its prediction …
Learning efficient task-dependent representations with synaptic plasticity
ColinBredenberg, EeroSimoncelli, CristinaSavin....
Published date-12/01/2020
Neural populations encode the sensory world imperfectly: their capacity is limited by the number of neurons, availability of metabolic and other biophysical resources, and intrinsic noise. The brain is presumably …