shagunsodhani

# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
from dataclasses import dataclass
from typing import Any

from omegaconf import MISSING, DictConfig, OmegaConf

import hydra
from hydra.core.config_store import ConfigStore
from hydra.types import ObjectConf

shagunsodhani

# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
from dataclasses import dataclass
from typing import Any

from omegaconf import MISSING, DictConfig, OmegaConf

import hydra
from hydra.core.config_store import ConfigStore
from hydra.types import ObjectConf

shagunsodhani

./lgtm.yml
 
  1:1       warning  missing document start "---"  (document-start)
 
  2:1       error    wrong indentation: expected 2 but found 0  (indentation)
 

 
./.pre-commit-config.yaml
 

shagunsodhani

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

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

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

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

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

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

shagunsodhani

PPDB: The Paraphrase Database

Introduction

• The paper presents a database of ranked English and Spanish paraphrases derived by:
  • Extracting lexical, phrasal, and syntactic paraphrases from large bilingual parallel corpora.
  • Computing the similarity scores for the pair of paraphrases using Google ngrams and the Annotated Gigaword corpus.
• Link to the paper

Extracting Paraphrase from Bilingual Text