AdCo is a contrastive-learning based self-supervised learning methods.

Copyright (C) 2020 Qianjiang*, Xiao Wang*, Wei Hu, Guo-Jun Qi

License: MIT for academic use.

Contact: Guo-Jun Qi ([email protected])

Contrastive learning relies on constructing a collection of negative examples that are sufficiently hard to discriminate against positive queries when their representations are self-trained. Existing contrastive learning methods either maintain a queue of negative samples over minibatches while only a small portion of them are updated in an iteration, or only use the other examples from the current minibatch as negatives. They could not closely track the change of the learned representation over iterations by updating the entire queue as a whole, or discard the useful information from the past minibatches. Alternatively, we present to directly learn a set of negative adversaries playing against the self-trained representation. Two players, the representation network and negative adversaries, are alternately updated to obtain the most challenging negative examples against which the representation of positive queries will be trained to discriminate. We further show that the negative adversaries are updated towards a weighted combination of positive queries by maximizing the adversarial contrastive loss, thereby allowing them to closely track the change of representations over time. Experiment results demonstrate the proposed Adversarial Contrastive (AdCo) model not only achieves superior performances (a top-1 accuracy of 73.2% over 200 epochs and 75.7% over 800 epochs with linear evaluation on ImageNet), but also can be pre-trained more efficiently with much shorter GPU time and fewer epochs.

CUDA version should be 10.1 or higher.

git clone [email protected]:maple-research-lab/AdCo.git && cd AdCo

You have two options to install dependency on your computer:

pip install -r requirements.txt --user

If you encounter any errors, you can install each library one by one:

pip install torch==1.7.1
pip install torchvision==0.8.2
pip install numpy==1.19.5
pip install Pillow==5.1.0
pip install tensorboard==1.14.0
pip install tensorboardX==1.7

conda create -n AdCo python=3.6.9
conda activate AdCo
pip install -r requirements.txt 

Each time when you want to run my code, simply activate the environment by

conda activate AdCo
conda deactivate(If you want to exit) 

This implementation only supports multi-gpu, DistributedDataParallel training, which is faster and simpler; single-gpu or DataParallel training is not supported.

python3 main_adco.py --sym=0 --lr=0.03 --memory_lr=3 --moco_t=0.12 --mem_t=0.02 --data=./datasets/imagenet2012 --dist_url=tcp://localhost:10001 

python3 main_adco.py --sym=1 --lr=0.03 --memory_lr=3 --moco_t=0.12 --mem_t=0.02 --data=./datasets/imagenet2012 --dist_url=tcp://localhost:10001

# e.g., training with 8192 negative samples and symmetrical loss
python3 main_adco.py --sym=1 --lr=0.04 --memory_lr=3 --moco_t=0.14 --mem_t=0.03 --cluster 8192 --data=./datasets/imagenet2012 --dist_url=tcp://localhost:10001

python3 main_adco.py --multi_crop=1 --lr=0.03 --memory_lr=3 --moco_t=0.12 --mem_t=0.02 --data=./datasets/imagenet2012 --dist_url=tcp://localhost:10001

So far we have yet to support multi crop with symmetrical loss, the feature will be added in future.

With a pre-trained model, we can easily evaluate its performance on ImageNet with:

python3 lincls.py --data=./datasets/imagenet2012 --dist-url=tcp://localhost:10001 --pretrained=input.pth.tar

Performance:

Really sorry that we can't provide 800 epochs' model, which is because that we train it with 32 internal GPUs and we can't download it because of company regulations. For downstream tasks, we found multi-200epoch model also had similar performance. Thus, we suggested you to use this model for downstream purposes.

Performance with different negative samples:

The performance is obtained on a single machine with 8*V100 GPUs.

cd VOC_CLF
python3 main.py --data=../datasets/voc --pretrained=../input.pth.tar

Here VOC directory should be the directory includes "vockit" directory.

python3 lincls.py --dataset=Place205 --sgdr=1 --data=./datasets/places205 --lr=5 --dist-url=tcp://localhost:10001 --pretrained=input.pth.tar

1. Install detectron2.

2. Convert a pre-trained AdCo model to detectron2's format:

   # in detection folder
   python3 convert-pretrain-to-detectron2.py input.pth.tar output.pkl
   

3. download VOC Dataset and COCO Dataset under "./detection/datasets" directory, following the directory structure requried by detectron2.

4. Run training:

   cd detection
   python train_net.py --config-file configs/pascal_voc_R_50_C4_24k_adco.yaml \
    --num-gpus 8 MODEL.WEIGHTS ./output.pkl
   

AdCo: Adversarial Contrast for Efficient Learning of Unsupervised Representations from Self-Trained Negative Adversaries

@article{hu2020adco,
  title={AdCo: Adversarial Contrast for Efficient Learning of Unsupervised Representations from Self-Trained Negative Adversaries},
  author={Hu, Qianjiang and Wang, Xiao and Hu, Wei and Qi, Guo-Jun},
  journal={arXiv preprint arXiv:2011.08435},
  year={2020}
}