This repo contains different kinds of presentations that I was giving. Most of them are just paper presentations for a reading group (disclaimer: not all of them are good).
Papers covered (sorry for arxiv links instead of conferences ones):
- Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One
- FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
- Principled Weight Initialization for Hypernetworks
- Multiplicative Interactions and Where to Find Them
- SinGAN: Learning a Generative Model from a Single Natural Image
- Normalized and Geometry-Aware Self-Attention Network for Image Captioning
- XLNet: Generalized Autoregressive Pretraining for Language Understanding
- Large Scale GAN Training for High Fidelity Natural Image Synthesis
- Continual learning with hypernetworks
- Learning deep representations by mutual information estimation and maximization
- On the "steerability" of generative adversarial networks
- Non-Adversarial Image Synthesis with Generative Latent Nearest Neighbors
- What's Hidden in a Randomly Weighted Neural Network?
- Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
- Learning Implicit Generative Models by Matching Perceptual Features
- Learning Implicit Fields for Generative Shape Modeling
- Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO
- Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
- Normalized and Geometry-Aware Self-Attention Network for Image Captioning
- Pruning neural networks without any data by iteratively conserving synaptic flow
- The State of Sparsity in Deep Neural Networks
- Weight Agnostic Neural Networks
- ReZero is All You Need: Fast Convergence at Large Depth