Search Results for author:
Found 10 papers, 4 papers with code
Affordance detection with Dynamic-Tree Capsule Networks
In the experimental evaluation we will show that our algorithm is superior to current affordance detection methods when faced with grasping previously unseen objects thanks to our Capsule Network enforcing a parts-to-whole representation.
Improving the Trainability of Deep Neural Networks through Layerwise Batch-Entropy Regularization
We show empirically that we can therefore train a "vanilla" fully connected network and convolutional neural network -- no skip connections, batch normalization, dropout, or any other architectural tweak -- with 500 layers by simply adding the batch-entropy regularization term to the loss function.
Continual Learning from Demonstration of Robotics Skills
We empirically demonstrate the effectiveness of this approach in remembering long sequences of trajectory learning tasks without the need to store any data from past demonstrations.
Momentum Capsule Networks
MoCapsNets are inspired by Momentum ResNets, a type of network that applies reversible residual building blocks.
Arguments for the Unsuitability of Convolutional Neural Networks for Non--Local Tasks
Convolutional neural networks have established themselves over the past years as the state of the art method for image classification, and for many datasets, they even surpass humans in categorizing images.
SU(3) analysis of four-quark operators: $K\toππ$ and vacuum matrix elements
Hadronic matrix elements of local four-quark operators play a central role in non-leptonic kaon decays, while vacuum matrix elements involving the same kind of operators appear in inclusive dispersion relations, such as those relevant in $\tau$-decay analyses.
High Energy Physics - Phenomenology High Energy Physics - Lattice
The two-loop perturbative correction to the (g-2)$_μ$ HLbL at short distances
The short-distance behaviour of the hadronic light-by-light (HLbL) contribution to $(g-2)_{\mu}$ has recently been studied by means of an operator product expansion in a background electromagnetic field.
High Energy Physics - Phenomenology
Evaluating the Progress of Deep Learning for Visual Relational Concepts
Convolutional Neural Networks (CNNs) have become the state of the art method for image classification in the last ten years.
25 years of CNNs: Can we compare to human abstraction capabilities?
We try to determine the progress made by convolutional neural networks over the past 25 years in classifying images into abstractc lasses.
Proceedings of the 37th Annual Workshop of the Austrian Association for Pattern Recognition (ÖAGM/AAPR), 2013
This volume represents the proceedings of the 37th Annual Workshop of the Austrian Association for Pattern Recognition (\"OAGM/AAPR), held May 23-24, 2013, in Innsbruck, Austria.
