Home /
Research
Showing 553 - 558 / 897
Investigating Learning in Deep Neural Networks using Layer-Wise Weight Change
AyushManishAgrawal, AtharvaTendle, HarshvardhanSikka....
Published date-11/13/2020
Understanding the per-layer learning dynamics of deep neural networks is of significant interest as it may provide insights into how neural networks learn and the potential for better training regimens. …
Efficient Subspace Search in Data Streams
EdouardFouché, FlorianKalinke, KlemensBöhm....
Published date-11/13/2020
OutlierDetection
In the real world, data streams are ubiquitous -- think of network traffic or sensor data. Mining patterns, e.g., outliers or clusters, from such data must take place in real …
Wisdom of the Ensemble: Improving Consistency of Deep Learning Models
LijingWang, DipanjanGhosh, MariaTeresaGonzalezDiaz....
Published date-11/13/2020
Deep learning classifiers are assisting humans in making decisions and hence the user's trust in these models is of paramount importance. Trust is often a function of constant behavior. From …
Multi-layered tensor networks for image classification
RaghavendraSelvan, SilasØrting, ErikBDam....
Published date-11/13/2020
ImageClassification, TensorNetworks
The recently introduced locally orderless tensor network (LoTeNet) for supervised image classification uses matrix product state (MPS) operations on grids of transformed image patches. The resulting patch representations are combined …
Image Animation with Perturbed Masks
YoavShalev, LiorWolf....
Published date-11/13/2020
ImageAnimation
We present a novel approach for image-animation of a source image by a driving video, both depicting the same type of object. We do not assume the existence of pose …
Testing the Genomic Bottleneck Hypothesis in Hebbian Meta-Learning
RasmusBergPalm, EliasNajarro, SebastianRisi....
Published date-11/13/2020
Meta-Learning
Recent work has shown promising results using Hebbian meta-learning to solve hard reinforcement learning problems and adapt-to a limited degree-to changes in the environment. In previous works each synapse has …