Paper reading notes. Only notes on checked papers are carefully written. This repo also includes other papers that are not listed here. I actually gitignores the pdf files so some directories are empty.
- Auto-Encoding Variational Bayes (half)
- Importance Weighted Autoencoders
- Weight Uncertainty in Neural Networks
- Categorical Reparameterization with Gumbel-Softmax
- beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework
- VRNN: A Recurrent Latent Variable Model for Sequential Data
- Towards Conceptual Compression
- Attend, Infer, Repeat: Fast Scene Understanding with Generative Models
- Towards a Neural Statistician
- Variational Continual Learning
- Multi-Object Representation Learning with Iterative Variational Inference
- SQAIR: Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objects
- Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks
- Multi-Object Representation Learning with Iterative Variational Inference
- Generative Temporal Models with Spatial Memory for Partially Observed Environments
- Neural Scene Representation and Rendering (+Supplementary Material)
- Tracking by Animation: Unsupervised Learning of Multi-Object Attentive Trackers
- Structured Inference Networks for Nonlinear State Space Models
- Auto-Encoding Sequential Monte Carlo
- Deep Variational Reinforcement Learning for POMDPs
- Neural Discrete Representation Learning
- Neural Relational Inference for Interacting Systems
- Amortized Bayesian Meta-Learning
- Variational Information Maximizing Exploration
- Variational Memory Addressing in Generative Models
- The Kanerva Machine: A Generative Distributed Memory
- Learning Latent Dynamics for Planning from Pixels
-
NEM/RNEM
-
GENESIS
-
Spatial Broadcast Decoder
-
Semi-supervised Learning with Deep Generative Models
-
Variational Inference with Normalizing Flows
-
DRAW: A Recurrent Neural Network For Image Generation
-
Neural Variational Inference and Learning
-
A simple neural network module for relational reasoning
-
MONet: Unsupervised Scene Decomposition and Representation
-
REASONING ABOUT PHYSICAL INTERACTIONS WITH OBJECT-ORIENTED PREDICTION AND PLANNING
-
Learning to Decompose and Disentangle Representations for Video Prediction
-
Avoiding Latent Variable Collapse with Generative Skip Models
-
MADE: Masked Autoencoder for Distribution Estimation
-
How to train deep variational autoencoders and probabilistic ladder networks.
-
Gradient Estimation Using Stochastic Computation Graphs
-
Stochastic backpropagation and approximate inference in deep generative models.
-
Deep Convolutional Inverse Graphics Network
-
Understanding Disentanglement in beta-VAE
-
Deep autoregressive network
-
Hierarchical variational models.
-
Neural Turing Machine
-
Capsule Net
-
A note on the evaluation of generative models
-
Black box variational inference.
-
Variational Inference for Monte Carlo Objectives
-
Stacked Capsule Autoencoders
-
Tigher Variational bounds are not necessarily better
-
Learning deep representations by mutual information estimation and maximization
-
Unsupervised State Representation Learning in Atari
-
Understanding posterior collapse in VAE
-
Isolating sources of disentanglemnt in VAE
-
Variational memory addressing in generative models
-
Lifelong Generative Modeling
-
unsupervised learning of object key points for perception and control
-
Memorization in Overparameterized Autoencoders
-
Neural ODEs
-
Why GAN can generate real images while VAE cannot?
-
Csgnet: Neural shape parser for constructive solid geometry
-
Peeking into the Future: Predicting Future Person Activities and Locations in Videos
-
Neural Processes
-
COPHY: C OUNTERFACTUAL L EARNING OF P HYSICAL DYNAMICS
-
Variational Inference: A Review for Statisticians
-
Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks
-
CLEVRER: CoLlision Events for Video REpresentation and Reasoning
-
LEARNING TO DESCRIBE SCENES WITH PROGRAMS
-
Galileo: Perceiving Physical Object Properties by Integrating a Physics Engine with Deep Learning
-
Activity forecasting
-
Neural scene de-rendering
-
DensePhysNet: Learning Dense Physical Object Representations via Multi-step Dynamic Interactions
-
Physics 101: Learning Physical Object Properties from Unlabeled Videos
-
A compositional object-based approach to learning physical dynamics (NPE)
-
STOCHASTIC PREDICTION OF MULTI-AGENT INTERACTIONS FROM PARTIAL OBSERVATIONS
-
UNSUPERVISED DISCOVERY OF PARTS, STRUCTURE, AND DYNAMICS
-
Unsupervised Learning of Latent Physical Properties Using Perception-Prediction Networks
-
A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning
-
Neural relational inference for interacting systems
-
Deep State-Space Models in Multi-Agent Systems
-
Fixing a broken ELBO
-
Interaction networks for learning about objects, relations and physics.
-
Compositional Video Prediction
-
Object-centric Forward Modeling for Model Predictive Control
-
Advances in variational inference
-
Visual interaction networks: Learning a physics simulator from video
-
Data-efficient model-based rl through unsupervised object discovery and curiosity-driven exploration
-
Faster attend-infer-repeat with tractable probabilistic models
-
Graph neural networks: A review of methods and applications.
-
RSQAIR
-
Factorized-VAE
-
Multivariate Time Series Imputation with Variational Autoencoders
-
STOCHASTIC VARIATIONAL VIDEO PREDICTION
-
Stochastic video generation with a learned prior.
-
Disentangled Sequential Autoencoder
-
THE VARIATIONAL GAUSSIAN PROCESS
-
Black Box Variational Inference
-
Tighter Variational Bounds are Not Necessarily Better
-
Isolating Sources of Disentanglement in Variational Autoencoders
-
REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models
-
On the Importance of Learning Aggregate Posteriors inMultimodal Variational Autoencoders
-
Invertible Residual Networks
-
Residual Flows for Invertible Generative Modeling
-
Explaining Image Classifiers by Counterfactual Generation
-
Isolating Souces of Disentanglement in Variational Autoencoders
-
Reinterpreting importance-weighted autoencoders
-
Inference Suboptimality in Variational Autoencoders
-
Doubly Reparameterized Gradient Estimators for Monte Carlo Objectives
-
#### Design Motifs for Probabilistic Generative Design
-
Continual Unsupervised Representation Learning
-
Mathematical Reasoning in Latent Space
-
Gated Variational AutoEncoders: Incorporating Weak Supervision to Encourage Disentanglement.
-
Unreproducible Research is Reproducible
-
An exploration of dropout with LSTMs
-
On Causal and Anticausal Learning
-
End-to-End Robotic Reinforcement Learning without Reward Engineering
-
Is a Good Representation Sufficient for Sample Efficient Reinforcement Learning?
-
From Reinforcement Learning to Deep Reinforcement Learning: An Overview
-
A Variational Inequality Perspective on Generative Adversarial Nets
-
Normalizing Flows for Probabilistic Modeling and Inference
-
Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations
-
A Primer in BERTology: What we know about how BERT works
-
Between MDPs and semi-MDPs: A Framework for Temporal Abstraction in Reinforcement Learning
-
Deep Infomax
-
DLRLSS, Yoshua Bengio, AGI and RNN
-
Don't Blame the Elbo! A Linear VAE Perspective on Posterior Collapse
-
Meta-Learning without Memorization
-
Unsupervised Learning of Object Keypoints for Perception and Control
-
VARIATIONAL STATE ENCODING AS INTRINSIC MOTIVATION IN REINFORCEMENT LEARNING
-
SuperDyna, Richard Sutton
-
Attention is all you need
-
InfoVAE: Information Maximizing Variational Autoencoders
-
Variational lossy encoder
-
Adversarial autoencoders
-
Hypernetwork
-
Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations
-
A Survey of Deep Reinforcement Learning in Video Games
-
Recurrent world models facilitates policy evaluation
-
A Closer Look at Memorization in Deep Networks
-
Hierarchical Decompositional Mixtures of Variational Autoencoders
-
Normalizing Flows for Probabilistic Modeling and Inference
-
On Learning Sets of Symmetric Elements
-
https://www.youtube.com/watch?v=Z56Jmr9Z34Q&list=PLyzOVJj3bHQuloKGG59rS43e29ro7I57J&index=2&t=0s
-
Disentangling the independently controllable factors of variation by interacting with the world
-
PolyGen: An Autoregressive Generative Model of 3D Meshes
-
Learning deep representations by mutual information estimation and maximization
-
BlockGAN
-
Inverse Graphics GAN
-
Self-Tuning Deep Reinforcement Learning
-
A review on generative adversarial networks: algorithms, theory and applications
-
Adversarial autoencoder
-
Stochastic Video Generation with a Learned Prior
-
A Primer in BERTology: What we know about how BERT works
-
Computing Machinery and Intelligence
-
Information generation as a functional basis of consciousness
-
RIM
-
Variational Temporal Abstraction
-
On Catastrophic Interference in Atari 2600 Games
-
Few-Shot Learning via Learning the Representation, Provably
-
Semantic Image Synthesis with SPADE
-
The importance of transparency and reproducibility in artificial intelligence research
-
Supervised autoencoders: Improving generalization performance with unsupervised regularizers
-
StyleGAN2 Distillation for Feed-forward Image Manipulation
-
Transformation-based Adversarial Video Prediction on Large-Scale Data
-
Google search: using GAN for representation learning
-
Object-based unsupervised scene representation learning
-
基于对象的无监督场景表征学习
-
Object-based Representation Learning from Unlabeled Videos
-
Object-Level Representation Learning for Few-Shot Image Classification
-
BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images
-
HoloGAN: Unsupervised learning of 3D representations from natural images
-
Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
-
Object-Oriented Dynamics Predictor
-
Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations
-
Out-of-Distribution Generalization via Risk Extrapolation (REx)
-
Stochastic Neural Network with Kronecker Flow
-
Planning in Dynamic Environments with Conditional Autoregressive Models
-
A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
-
Ordered Memory
-
Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
-
Gauge Equivariant Mesh CNNs: Anisotropic convolutions on geometric graphs
-
AutoML-Zero: Evolving Machine Learning Algorithms From Scratch
-
VQ-DRAW
-
Gauge Equivariant GCNN
-
Semi-supervised StyleGAN for disentanglement learning
-
On Catastrophic Interference in Atari 2600 Games
-
Deep Meditations: Controlled navigation of latent space
-
Deep SDF
-
A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
-
Representation Learning Through Latent Canonicalizations
-
Deep Sets for Generalization in RL
-
Disentangled Relational Representations for Explaining and Learning from Demonstration
-
Indirectly Encoded Sodarace for Artificial Life
-
SUMO: Unbiased Estimation of Log Marginal Probability for Latent Variable Models. (arXiv:2004.00353v1 [cs.LG]) https://ift.tt/2UWC9Ue
-
A theory of independent mechanisms for extrapolation in generative models. (arXiv:2004.00184v1 [cs.LG])
-
Use adversarial learning in the latent space. This way you don't need to produce perfect images.
-
Self-supervised Object-centric Representations for Reinforcement Learning
-
Construction as the Manipulation of Structure
-
Learning to Control Visual Abstractions for Structured Exploration in Deep Reinforcement Learning
-
Challenging common assumptions in the unsupervised learning of disentangled representations
-
Unsupervised Learning of Probably Symmetric Deformable 3D Objects in the Wild
-
Symmetry-Based Disentangled Representation Learning requires Interaction with Environments
-
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
-
Convolution and attention (transformer)
-
[Social GAN](Social GAN)
-
Temporal Generalization, long term dependency
-
HoloGAN
-
The equivalence between Stein variational gradient descent and black-box variational inference
-
Learning to Manipulate Individual Objects in an Image
-
pytorch-lightning
-
DeepSDF
-
NeRF
-
Deep (Neural) Volume Rendering