Design of potent antimalarials with generative chemistry

WJ Godinez, EJ Ma, AT Chao, L Pei… - Nature Machine …, 2022 - nature.com
Recent advances in generative modelling allow designing novel compounds through deep
neural networks. One such neural network model, JT-VAE (the Junction Tree Variational …

Feature space saturation during training

ML Richter, J Shenk, W Byttner, A Arpteg… - arxiv preprint arxiv …, 2020 - arxiv.org
We propose layer saturation-a simple, online-computable method for analyzing the
information processing in neural networks. First, we show that a layer's output can be …

Complexity for deep neural networks and other characteristics of deep feature representations

RA Janik, P Witaszczyk - arxiv preprint arxiv:2006.04791, 2020 - arxiv.org
We define a notion of complexity, which quantifies the nonlinearity of the computation of a
neural network, as well as a complementary measure of the effective dimension of feature …

[PDF][PDF] Delve: Neural Network Feature Variance Analysis

J Shenk, ML Richter, W Byttner - Journal of Open Source Software, 2022 - joss.theoj.org
Designing neural networks is a complex task. Deep neural networks are often referred to as
“black box” models-little insight in the function they approximate is gained from looking at the …

Exploring the properties and evolution of neural network eigenspaces during training

ML Richter, L Malihi, AKP Windler… - … on Machine Vision …, 2022 - ieeexplore.ieee.org
We investigate properties and the evolution of the emergent inference process inside neural
networks using layer saturation [1] and logistic regression probes [2]. We demonstrate that …

Analyzing the Inference Process in Deep Convolutional Neural Networks using Principal Eigenfeatures, Saturation and Logistic Regression Probes

ML Richter, L Malihi, AKP Windler… - Journal of Applied …, 2023 - jaree.scu.ac.ir
The predictive performance of a neural network depends on the one hand on the difficulty of
a problem, defined by the number of classes and complexity of the visual domain, and on …

Investigating the Learning Progress of CNNs in Script Identification Using Gradient Values

E Tomioka, K Morita, NC Shirai… - 2019 4th …, 2019 - ieeexplore.ieee.org
Demands for an automatic translation based on Camera-based Multilingual Optical
Character Recognition (CM-OCR) are increasing. In addition, CM-OCR methods usually …

[PDF][PDF] CNN を用いた言語判定における学習過程の分析

冨岡永伍 - mie-u.repo.nii.ac.jp
社会の国際化に伴い, カメラベース OCR を用いた自動翻訳ソフトウェアの需要が高まっている.
撮影した画像から多言語の文字を同時に直接認識するカメラベース OCR の実現は困難であるため …

[CYTOWANIE][C] TOWARDS GUIDED NEURAL ARCHITECTURE DESIGN BY SPECTRAL ANALYSIS