Continuous Query Decomposition for Complex Query Answering in Incomplete Knowledge Graphs

Last update: Dec 29, 2022

Related tags

Overview

Continuous Query Decomposition

This repository contains the official implementation for our ICLR 2021 (Oral) paper, Complex Query Answering with Neural Link Predictors:

@inproceedings{
    arakelyan2021complex,
    title={Complex Query Answering with Neural Link Predictors},
    author={Erik Arakelyan and Daniel Daza and Pasquale Minervini and Michael Cochez},
    booktitle={International Conference on Learning Representations},
    year={2021},
    url={https://openreview.net/forum?id=Mos9F9kDwkz}
}

In this work we present CQD, a method that reuses a pretrained link predictor to answer complex queries, by scoring atom predicates independently and aggregating the scores via t-norms and t-conorms.

Our code is based on an implementation of ComplEx-N3 available here.

Please follow the instructions next to reproduce the results in our experiments.

1. Install the requirements

We recommend creating a new environment:

% conda create --name cqd python=3.8 && conda activate cqd
% pip install -r requirements.txt

2. Download the data

We use 3 knowledge graphs: FB15k, FB15k-237, and NELL. From the root of the repository, download and extract the files to obtain the folder data, containing the sets of triples and queries for each graph.

% wget http://data.neuralnoise.com/cqd-data.tgz
% tar xvf cqd-data.tgz

3. Download the models

Then you need neural link prediction models -- one for each of the datasets. Our pre-trained neural link prediction models are available here:

% wget http://data.neuralnoise.com/cqd-models.tgz
% tar xvf cqd-data.tgz

3. Alternative -- Train your own models

To obtain entity and relation embeddings, we use ComplEx. Use the next commands to train the embeddings for each dataset.

FB15k

% python -m kbc.learn data/FB15k --rank 1000 --reg 0.01 --max_epochs 100  --batch_size 100

FB15k-237

% python -m kbc.learn data/FB15k-237 --rank 1000 --reg 0.05 --max_epochs 100  --batch_size 1000

NELL

% python -m kbc.learn data/NELL --rank 1000 --reg 0.05 --max_epochs 100  --batch_size 1000

Once training is done, the models will be saved in the models directory.

4. Answering queries with CQD

CQD can answer complex queries via continuous (CQD-CO) or combinatorial optimisation (CQD-Beam).

CQD-Beam

Use the kbc.cqd_beam script to answer queries, providing the path to the dataset, and the saved link predictor trained in the previous step. For example,

% python -m kbc.cqd_beam --model_path models/[model_filename].pt

Example:

% PYTHONPATH=. python3 kbc/cqd_beam.py \
  --model_path models/FB15k-model-rank-1000-epoch-100-*.pt \
  --dataset FB15K --mode test --t_norm product --candidates 64 \
  --scores_normalize 0 data/FB15k

models/FB15k-model-rank-1000-epoch-100-1602520745.pt FB15k product 64
ComplEx(
  (embeddings): ModuleList(
    (0): Embedding(14951, 2000, sparse=True)
    (1): Embedding(2690, 2000, sparse=True)
  )
)

[..]

This will save a series of JSON fils with results, e.g.

% cat "topk_d=FB15k_t=product_e=2_2_rank=1000_k=64_sn=0.json"
{
  "MRRm_new": 0.7542805715523118,
  "MRm_new": 50.71081983144581,
  "[email protected]_new": 0.6896709378392843,
  "[email protected]_new": 0.7955001359095913,
  "[email protected]_new": 0.8676865172456019
}

CQD-CO

Use the kbc.cqd_co script to answer queries, providing the path to the dataset, and the saved link predictor trained in the previous step. For example,

% python -m kbc.cqd_co data/FB15k --model_path models/[model_filename].pt --chain_type 1_2

Final Results

All results from the paper can be produced as follows:

% cd results/topk
% ../topk-parse.py *.json | grep rank=1000
d=FB15K rank=1000 & 0.779 & 0.584 & 0.796 & 0.837 & 0.377 & 0.658 & 0.839 & 0.355
d=FB237 rank=1000 & 0.279 & 0.219 & 0.352 & 0.457 & 0.129 & 0.249 & 0.284 & 0.128
d=NELL rank=1000 & 0.343 & 0.297 & 0.410 & 0.529 & 0.168 & 0.283 & 0.536 & 0.157
% cd ../cont
% ../cont-parse.py *.json | grep rank=1000
d=FB15k rank=1000 & 0.454 & 0.191 & 0.796 & 0.837 & 0.336 & 0.513 & 0.816 & 0.319
d=FB15k-237 rank=1000 & 0.213 & 0.131 & 0.352 & 0.457 & 0.146 & 0.222 & 0.281 & 0.132
d=NELL rank=1000 & 0.265 & 0.220 & 0.410 & 0.529 & 0.196 & 0.302 & 0.531 & 0.194

Continuous Query Decomposition for Complex Query Answering in Incomplete Knowledge Graphs

Related tags

Overview

Continuous Query Decomposition

1. Install the requirements

2. Download the data

3. Download the models

3. Alternative -- Train your own models

FB15k

FB15k-237

NELL

4. Answering queries with CQD

CQD-Beam

CQD-CO

Final Results

Owner

UCL Natural Language Processing

Hierarchical Aggregation for 3D Instance Segmentation (ICCV 2021)

[NeurIPS 2021] Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training

Like a cowsay but without cows!

A collection of SOTA Image Classification Models in PyTorch

Bounding Wasserstein distance with couplings

The Official PyTorch Implementation of DiscoBox.

UNAVOIDS: Unsupervised and Nonparametric Approach for Visualizing Outliers and Invariant Detection Scoring

implementation of the paper "MarginGAN: Adversarial Training in Semi-Supervised Learning"

Code for "On Memorization in Probabilistic Deep Generative Models"

Official code for "Mean Shift for Self-Supervised Learning"

DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.

Semi-supervised Implicit Scene Completion from Sparse LiDAR

Radar-to-Lidar: Heterogeneous Place Recognition via Joint Learning

ICCV2021 - Mining Contextual Information Beyond Image for Semantic Segmentation

Detecting Human-Object Interactions with Object-Guided Cross-Modal Calibrated Semantics

Fast, modular reference implementation and easy training of Semantic Segmentation algorithms in PyTorch.

Face Recognition and Emotion Detector Device

Jax/Flax implementation of Variational-DiffWave.

Source code of the paper "Deep Learning of Latent Variable Models for Industrial Process Monitoring".

improvement of CLIP features over the traditional resnet features on the visual question answering, image captioning, navigation and visual entailment tasks.