Papers I Read Notes and Summaries

The Lottery Ticket Hypothesis - Training Pruned Neural Networks

Introduction

  • Empirical evidence indicates that at training time, the neural networks need to be of significantly larger... Continue reading


Cyclical Learning Rates for Training Neural Networks

Introduction

  • Conventional wisdom says that when training neural networks, learning rate should monotonically decrease. This insight forms... Continue reading


Improving Information Extraction by Acquiring External Evidence with Reinforcement Learning

Introduction

  • Information Extraction - Given a query to be answered and an external search engine, information extraction... Continue reading


An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks

Introduction

  • Catastrophic Forgetting refers to the phenomenon where when a learning system is trained on two tasks... Continue reading


Learning an SAT Solver from Single-Bit Supervision

Introduction

  • The paper presents NeuroSAT, a message passing neural network that is trained to predict if a... Continue reading


Neural Relational Inference for Interacting Systems

Introduction

  • The paper presents Neural Relational Inference (NRI) model which can infer underlying interactions in a dynamical... Continue reading


Stylistic Transfer in Natural Language Generation Systems Using Recurrent Neural Networks

Introduction


Get To The Point - Summarization with Pointer-Generator Networks

Introduction


StarSpace - Embed All The Things!

Introduction

  • The paper describes a general purpose neural embedding model where different type of entities (described in... Continue reading


Emotional Chatting Machine - Emotional Conversation Generation with Internal and External Memory

  • The paper proposes ECM (Emotional Chatting Machine) which can generate both semantically and emotionally appropriate responses in a... Continue reading