Yizheng Chen, Zhoujie Ding, and David Wagner
Paper: https://arxiv.org/abs/2302.04332
Download this from Google Drive. The zipped file contains DREBIN features of the APIGraph dataset and AndroZoo dataset we used in the paper.
Extract the downloaded file to data/, such that the datasets are under data/gen_apigraph_drebin and data/gen_androzoo_drebin.
- We collected
data/gen_apigraph_drebinby downloading the sample hashes released by the APIGraph paper. The samples are from 2012 to 2018. - We collected
data/gen_androzoo_drebinby downloading apps from AndroZoo. The samples are from 2019 to 2021.
The following example trains an enc-mlp model using hi-dist-xent loss, i.e., our Hierarchical Contrastive Classifier, and it runs active learning with 200 samples / month budget using our Psuedo Loss Sample Selector.
#! /bin/bash
SEQ=088
LR=0.003
OPT=sgd
SCH=step
DECAY=0.95
E=250
WLR=0.00015
WE=100
DATA=gen_apigraph_drebin
TRAIN_START=2012-01
TRAIN_END=2012-12
TEST_START=2013-01
TEST_END=2018-12
RESULT_DIR=results_ours
AL_OPT=adam
CNT=200
modeldim="512-384-256-128"
S='half'
B=1024
LOSS='hi-dist-xent'
TS=$(date "+%m.%d-%H.%M.%S")
nohup python -u relabel.py \
--data ${DATA} \
--benign_zero \
--mdate 20230501 \
--train_start ${TRAIN_START} \
--train_end ${TRAIN_END} \
--test_start ${TEST_START} \
--test_end ${TEST_END} \
--encoder simple-enc-mlp \
--classifier simple-enc-mlp \
--loss_func ${LOSS} \
--enc-hidden ${modeldim} \
--mlp-hidden 100-100 \
--mlp-dropout 0.2 \
--sampler ${S} \
--bsize ${B} \
--optimizer ${OPT} \
--scheduler ${SCH} \
--learning_rate ${LR} \
--lr_decay_rate ${DECAY} \
--lr_decay_epochs "10,500,10" \
--epochs ${E} \
--encoder-retrain \
--al_optimizer ${AL_OPT} \
--warm_learning_rate ${WLR} \
--al_epochs ${WE} \
--xent-lambda 100 \
--display-interval 180 \
--al \
--count ${CNT} \
--local_pseudo_loss \
--reduce "none" \
--sample_reduce 'mean' \
--result experiments/020_revision/${RESULT_DIR}/gen_apigraph_cnt${CNT}_${SEQ}_warm_lr${LR}_${OPT}_${SCH}_${DECAY}_e${E}_${AL_OPT}_wlr${WLR}_we${WE}_test_${TEST_START}_${TEST_END}_cnt${CNT}.csv \
--log_path experiments/020_revision/${RESULT_DIR}/gen_apigraph_cnt${CNT}_${SEQ}_warm_lr${LR}_${OPT}_${SCH}_${DECAY}_e${E}_${AL_OPT}_wlr${WLR}_we${WE}_test_${TEST_START}_${TEST_END}_cnt${CNT}_${TS}.log \
>> experiments/020_revision/${RESULT_DIR}/gen_apigraph_cnt${CNT}_${SEQ}_warm_lr${LR}_${OPT}_${SCH}_${DECAY}_e${E}_${AL_OPT}_wlr${WLR}_we${WE}_test_${TEST_START}_${TEST_END}_cnt${CNT}_${TS}.log 2>&1 &
We used the scripts under experiments/020_revision to run experiments in the paper. We ran these jobs on a Slurm GPU cluster (thanks to Center for AI Safety). If you would like to run the same script on a GPU server, not managed by Slurm, you would need to remove the lines starting with #SBATCH and also the last line (i.e. wait).
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent