ICLR 2018 accepted papers on Learning theory and algorithms
24 of the 337 papers accepted at ICLR 2018 (7.1%) are on Learning theory and algorithms, compared with 6.1% at ICLR 2017. They fall into 8 topics.
See ICLR 2018 in the interactive map
Submitting to ICLR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Kernels and regression theory · 7 papers
- A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks (296 citations)
- Sensitivity and Generalization in Neural Networks: an Empirical Study (285 citations)
- Demystifying MMD GANs (171 citations)
- On the Discrimination-Generalization Tradeoff in GANs (56 citations)
- Adaptive Dropout with Rademacher Complexity Regularization (21 citations)
- Initialization matters: Orthogonal Predictive State Recurrent Neural Networks (18 citations)
- Not-So-Random Features
Combinatorial and search optimization · 6 papers
- Neural-Guided Deductive Search for Real-Time Program Synthesis from Examples (70 citations)
- Measuring the Intrinsic Dimension of Objective Landscapes (59 citations)
- Towards Synthesizing Complex Programs From Input-Output Examples (31 citations)
- Hyperparameter optimization: a spectral approach (25 citations)
- Combining Symbolic Expressions and Black-box Function Evaluations in Neural Programs (9 citations)
- Divide and Conquer Networks (6 citations)
Prediction and decision losses · 4 papers
- Smooth Loss Functions for Deep Top-k Classification (70 citations)
- Global Optimality Conditions for Deep Neural Networks (52 citations)
- Critical Points of Linear Neural Networks: Analytical Forms and Landscape Properties (37 citations)
- SEARNN: Training RNNs with global-local losses (13 citations)
Matrix and tensor methods · 2 papers
Sample complexity · 2 papers
Submodular and game algorithms · 1 paper
Clustering · 1 paper
- SpectralNet: Spectral Clustering using Deep Neural Networks (115 citations)
Optimal transport · 1 paper
- Learning Wasserstein Embeddings (8 citations)
