A Variational Analysis of Stochastic Gradient Algorithms(ICML2016)
[paper]
[supplement]
An Information Theoretic Perspective on Multiple Classifier Systems [paper] アンサンブル
Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledge
Emmanuel de Bezenac, Arthur Pajot, Patrick Gallinari
[paper]
High-dimensional dynamics of generalization error in neural networks
Madhu S. Advani, Andrew M. Saxe
[paper]
Understanding Hidden Memories of Recurrent Neural Networks [paper]
How deep learning works --The geometry of deep learning
Xiao Dong, Jiasong Wu, Ling Zhou
[paper]
Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Step
William Fedus, Mihaela Rosca, Balaji Lakshminarayanan, Andrew M. Dai, Shakir Mohamed, Ian Goodfellow
[paper]
On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
Nitish Shirish Keskar, Dheevatsa Mudigere, Jorge Nocedal, Mikhail Smelyanskiy, Ping Tak Peter Tang
[paper]
Understanding Generalization and Stochastic Gradient Descent
Samuel L. Smith, Quoc V. Le
[paper]
Deep Neural Network Capacity
Aosen Wang, Hua Zhou, Wenyao Xu, Xin Chen
[paper]
Understanding Black-box Predictions via Influence Functions(ICML2017 best paper)
Pang Wei Koh, Percy Liang
[paper]
Methods for Interpreting and Understanding Deep Neural Networks(survey)
Grégoire Montavon, Wojciech Samek, Klaus-Robert Müller
[paper]
Spectrally-normalized margin bounds for neural networks
Peter Bartlett, Dylan J. Foster, Matus Telgarsky
[paper]
Exploring Generalization in Deep Learning
Behnam Neyshabur, Srinadh Bhojanapalli, David McAllester, Nathan Srebro
[paper]
A Closer Look at Memorization in Deep Networks
Devansh Arpit, Stanisław Jastrzębski, Nicolas Ballas, David Krueger, Emmanuel Bengio, Maxinder S. Kanwal, Tegan Maharaj, Asja Fischer, Aaron Courville, Yoshua Bengio, Simon Lacoste-Julien
[paper]
All-but-the-Top: Simple and Effective Postprocessing for Word Representations
Jiaqi Mu, Suma Bhat, Pramod Viswanath
[paper]
[ArXivtimes]
Why Regularized Auto-Encoders Learn Sparse Representation?(ICML2016)
[paper]
[supplementary]
nalysis of Deep Neural Networks with the Extended Data Jacobian Matrix(ICML2016)
[paper]
Benchmarking Deep Reinforcement Learning for Continuous Control(ICML2016)
[paper]
[supplementary]
Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units(ICML2016)
[paper]
An Analysis of the t-SNE Algorithm for Data Visualization (COLT2018)
Sanjeev Arora, Wei Hu, Pravesh K. Kothari
[paper]
Network Dissection: Quantifying Interpretability of Deep Visual Representations
David Bau, Bolei Zhou*, Aditya Khosla, Aude Oliva, Antonio Torralba*
[project]
Visualizing Residual Networks
Brian Chu, Daylen Yang, Ravi Tadinada
[paper]
Visualizing Deep Neural Network Decisions: Prediction Difference Analysis
Luisa M Zintgraf, Taco S Cohen, Tameem Adel, Max Welling
[paper]
Visualizing Data using t-SNE
[peper]
Understanding trained CNNs by indexing neuron selectivity
Ivet Rafegas, Maria Vanrell, Luis A. Alexandre
[paper]
Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit
Kevin Schawinski, Ce Zhang, Hantian Zhang, Lucas Fowler, Gokula Krishnan Santhanam
[paper]
DeepIU: An Architecture for Image Understanding(ACS2016)
[paper]
Symbolic and structural models for image understanding Part II: Ontologies and description logics(IJCAI tutorial)
[slide]
#language emergence
Emergence of Grounded Compositional Language in Multi-Agent Populations
Igor Mordatch, Pieter Abbeel
[paper]
[blog]
A Paradigm for Situated and Goal-Driven Language Learning
Jon Gauthier, Igor Mordatch
[paper]
Emergence of Gricean Maxims from Multi-Agent Decision Theory
[paper]