You can always learn a lot from the papers presented at NeurIPS
There is some good analysis already on the web.
From Chip Huygen – neurips 2019 analysis and from David Abel neurips 2019 analysis
Most major players were also well represented at NeurIPS including
For my students at University of Oxford #AI #Cloud #Edge I did an analysis of neurips papers by theme based on the neurips 2019 schedule. I found it easier to analyse papers based on theme
The themes covered in NeurIPS were
Algorithms
- Adaptive Data Analysis
- Adversarial Learning
- Bandit Algorithms
- Boosting and Ensemble Methods
- Clustering
- Components Analysis (e.g., CCA, ICA, LDA, PCA)
- Density Estimation
- Dynamical Systems
- Kernel Methods
- Meta-Learning
- Missing Data
- Model Selection and Structure Learning
- Regression
- Representation Learning
- Semi-Supervised Learning
- Similarity and Distance Learning
- Structured Prediction
- Uncertainty Estimation
- Unsupervised Learning
Applications
- Body Pose, Face, and Gesture Analysis
- Communication- or Memory-Bounded Learning
- Computer Vision
- Dialog- or Communication-Based Learning
- Game Playing
- Image Segmentation
- Object Detection
- Privacy, Anonymity, and Security
- Recommender Systems
- Robotics
- Web Applications and Internet Data
- Biologically Plausible Deep Networks
Deep Learning
- Deep Autoencoders
- Efficient Inference Methods
- Generative Models
- Interaction-Based Deep Networks
- Optimization for Deep Networks
- Predictive Models
- Recurrent Networks
- Visualization or Exposition Techniques for Deep Networks
Optimization
- Combinatorial Optimization
- Convex Optimization
- Non-Convex Optimization
- Stochastic Optimization
Probabilistic Methods
- Causal Inference
- Distributed Inference
- Gaussian Processes
- Hierarchical Models
- MCMC
- Variational Inference
Reinforcement Learning and Planning
- Decision and Control
- Exploration
- Markov Decision Processes
- Model-Based RL
- Multi-Agent RL
- Navigation
- Reinforcement Learning
Theory
- Computational Complexity
- Control Theory
- Frequentist Statistics
- Hardness of Learning and Approximations
- Learning Theory
A full list of papers by theme below
NeurIPS 2019 analysis of papers by theme
Image source: Yandex @neurips
