Question Answering with Subgraph Embeddings

Key Idea

Background

Task setup

Embedding Questions and Answers

Scoring function used in this approach
Embedding Matrix

Representing Candidate Answers

Training and Loss function

Inference

  • One possibility of this candidate set — is to take complete knowledge base, but has speed and precision issues. So authors decided to go ahead with generation of this answer candidate set for each question.
  • Authors went ahead with two approaches: C1 -> where they consider only the Freebase triples that have the question entity in them such that all 1-hop answer entities are included in the candidate set.
  • C2 -> In order to be able to answer answers which are at 2-hop, authors are leveraging beam-search to expand the 2nd hop relations candidates. The model first ranks relations type that are likely to be expressed in question and only add 2nd hop candidates to answer candidate set when these relations appear in those paths.

Experiments and Results

  • Authors have shared numbers in the table that indicates their strategy C2 performs the best in terms of F1 score and Precision@1. Replacing C2 by C1 shows a drop in performance as system is not able to handle any questions other than 1-hop ones.
  • Overall, the results presented in this paper verify the hypothesis that a richer representation for answers using local subgraph can store the pertinent information which helps in correctly answering the question.

--

--

--

kartikperisetla.github.io

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Towards Transferable Targeted Attack

3 silver bullets of word embeddings in NLP

Learning a five-in-a-row policy using PyTorch

6 Essential Tips to Improve Machine Learning

You cannot solve a Problem — if you are a Part of the Problem

Why is DVC Better Than Git and Git-LFS in Machine Learning Reproducibility

Demystifying Deep Deterministic Policy Gradient(DDPG) using ChainerRL and OpenAI-baselines

Chapter 2.2 : Self-Driving Car [Intro to TensorFlow & Deep Neural Network]

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Kartik Perisetla

Kartik Perisetla

kartikperisetla.github.io

More from Medium

Making the ‘Good Finds’ Happen!

Time to say goodbye to GDP? 5 alternative economic models (to our current neoclassical model)

Customers are your Currency

The development of JACKAL products is moving along smoothly and we are happy to formalize a…