Skip to content

Instantly share code, notes, and snippets.

Shagun Sodhani shagunsodhani

Block or report user

Report or block shagunsodhani

Hide content and notifications from this user.

Learn more about blocking users

Contact Support about this user’s behavior.

Learn more about reporting abuse

Report abuse
View GitHub Profile
@shagunsodhani
shagunsodhani / gist:bc62cc7ac9ad6876a0ca3c5fd86b2402
Last active May 16, 2018 — forked from asabaylus/gist:3071099
Github Markdown Heading Anchors
View gist:bc62cc7ac9ad6876a0ca3c5fd86b2402

Model Yes/No Number Other All


Baseline + count (ours) 80.76 47.3 55.99 64.16 Unidirectional LSTM 80.85 47.34 56.09 64.25 Use 20 objects 80.46 48.19 56.11 64.22 Embed. size 100 78.54 47.06 55.59 63.09 Confidence thresh. 0.2 80.43 48.15 56.06 64.18 Use second glimpse 80.23 48.02 56.08 64.1 Baseline (ours) 80.23 43.52 55.92 63.43 Baseline (bidirect. LSTM) 43.62 80.31 55.96 63.49

View issue-suggestion1.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View issue.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View cluster_setup.sh
#!/bin/bash
module use ~/projects/rpp-bengioy/modules/*/Core
#module load miniconda3
# for CEDAR
#conda create -n mila --clone="/project/6008064/software/2017/Core/miniconda3/4.3.27"
# for GRAHAM
conda create -n mila --clone="/project/6000167/anaconda3"
conda install pytorch torchvision -c pytorch
@shagunsodhani
shagunsodhani / Learning to Generate Reviews and Discovering Sentiment.md
Last active Jul 8, 2019
Notes for "Learning to Generate Reviews and Discovering Sentiment" paper
View Learning to Generate Reviews and Discovering Sentiment.md

Learning to Generate Reviews and Discovering Sentiment

Summary

The authors train a character-RNN (using mLSTM units) over Amazon Product Reviews (82 million reviews) and use the char-RNN as the feature extractor for sentiment analysis. These unsupervised features beat state of the art results for the dataset while are outperformed by supervised approaches on other datasets. Most important observation is that the authors find a single neuron (called as the sentiment neuron) which alone achieves a test accuracy of 92.3% thus giving the impression that the sentiment concept has been captured in that single neuron. Switching this neuron on (or off) during the generative process produces positive (or negative) reviews.

Notes

  • The paper aims to evaluate if the low level features captured by char-RNN can support learning of high-level representations.
@shagunsodhani
shagunsodhani / Seeing the Arrow of Time.md
Created Apr 10, 2017
Notes for paper "Seeing the Arrow of Time"
View Seeing the Arrow of Time.md

Seeing the Arrow of Time

Introduction

  • Given a video, can a machine learning system detect the arrow of time and distinguish whether the video is running forward or backwards.
  • Link to the paper

Datasets

  • Youtube Dataset
@shagunsodhani
shagunsodhani / End-to-end optimization of goal-driven and visually grounded dialogue systems.md
Created Apr 5, 2017
Notes for paper "End-to-end optimization of goal-driven and visually grounded dialogue systems"
View End-to-end optimization of goal-driven and visually grounded dialogue systems.md

End-to-end optimization of goal-driven and visually grounded dialogue systems

Introduction

  • The paper introduces an architecture for end-to-end Reinforcement Learning (RL) optimization for task-oriented dialogue systems and its application to a multimodal task - grounding the dialogue into a visual context.

Encoder Decoder Models vs RL Models

  • Encoder Decoder models do not account for the planning problems (which are inherent in the dialogue systems) and do not integrate seamlessly with external contexts or knowledge bases.
  • RL models can handle the planning problem but require online learning and a predefined structure of the task.
@shagunsodhani
shagunsodhani / GuessWhat.md
Created Mar 28, 2017
Summary of "GuessWhat?! Visual object discovery through multi-modal dialogue" paper
View GuessWhat.md

GuessWhat?! Visual object discovery through multi-modal dialogue

Introduction

  • The paper introduces GuessWhat - a two-player guessing game where the goal is to locate an object in a rich image scene.
  • The game is used to produce a large scale dataset of visual question-answer pairs on the image.
  • The paper also describes three tasks based on the game and provides a neural architecture based baselines for each task.
  • Link to the paper

GuessWhat?! Game

@shagunsodhani
shagunsodhani / Semantic Parsing via Paraphrasing.md
Created Mar 21, 2017
Notes for Semantic Parsing via Paraphrasing paper
View Semantic Parsing via Paraphrasing.md

Semantic Parsing via Paraphrasing

Introduction

  • Semantic parsing is the problem of mapping natural language utterances into logical forms that can be executed on a Knowledge Base (KB).
  • The paper presents a new approach to semantic parsing that uses paraphrasing to leverage the large amount of text which is not covered by the KB.
  • Link to the paper

Approach

@shagunsodhani
shagunsodhani / Traversing Knowledge Graphs in Vector Space.md
Created Mar 13, 2017
Notes for "Traversing Knowledge Graphs in Vector Space" paper
View Traversing Knowledge Graphs in Vector Space.md

Traversing Knowledge Graphs in Vector Space

Introduction

  • The paper describes a "compositional" training approach for vector space models, corresponding to Knowledge Bases (KBs).
  • The new approach improves the system's ability to answer path queries and impute missing information for the KBs.
  • Link to the paper

Task

You can’t perform that action at this time.