Home /
Research
Showing 25 - 30 / 904
Neural Pooling for Graph Neural Networks
Anonymous....
Published date-01/01/2021
GraphClassification
Tasks such as graph classification, require graph pooling to learn graph-level representations from constituent node representations. In this work, we propose two novel methods using fully connected neural network layers …
Bayesian Context Aggregation for Neural Processes
Anonymous....
Published date-01/01/2021
BayesianInference, Multi-TaskLearning
Formulating scalable probabilistic regression models with reliable uncertainty estimates has been a long-standing challenge in machine learning research. Recently, casting probabilistic regression as a multi-task learning problem in terms of …
Uniform Manifold Approximation with Two-phase Optimization
Anonymous....
Published date-01/01/2021
DimensionalityReduction
We present a dimensionality reduction algorithm called Uniform Manifold Approximation with Two-phase Optimization (UMATO), that aims to preserve both the global and local structures of high-dimensional data. Most existing dimensionality …
Ruminating Word Representations with Random Noise Masking
Anonymous....
Published date-01/01/2021
TextClassification, WordEmbeddings
We introduce a training method for better word representation and performance, which we call \textbf{GraVeR} (\textbf{Gra}dual \textbf{Ve}ctor \textbf{R}umination). The method is to gradually and iteratively add random noises and bias …
Structure and randomness in planning and reinforcement learning
Anonymous....
Published date-01/01/2021
Planning in large state spaces inevitably needs to balance depth and breadth of the search. It has a crucial impact on planners performance and most manage this interplay implicitly. We …
Gradient-based training of Gaussian Mixture Models for High-Dimensional Streaming Data
Anonymous....
Published date-01/01/2021
We present an approach for efficiently training Gaussian Mixture Models by SGD on non-stationary, high-dimensional streaming data. Our training scheme does not require data-driven parameter initialization (e.g., k-means) and has …