Thank you for your great work!
Since the UADA attack takes quite a long time, I was wondering if you have tried speeding it up by:
- increasing the learning rate
- increasing the batch size
- reducing the number of iterations
- reducing the inner loop
If so, did any of these work? Also, how sensitive was the performance to these settings?
Thank you in advance!
python roboticAttack/VLAAttacker/UADA_wrapper.py \
--maskidx 0 \
--lr 2e-3 \
--server $current_dir \
--device 0 \
--iter 2000 \
--accumulate 1 \
--bs 8 \
--warmup 20 \
--tags "debug testrun" \
--filterGripTrainTo1 false \
--geometry true \
--patch_size "3,50,50" \
--wandb_project "false" \
--wandb_project "false" \
--wandb_entity "xxx" \
--innerLoop 50 \
--dataset "libero_spatial" # "libero_spatial" / "libero_10" / "libero_goal" / "libero_goal" / "bridge_orig"
Thank you for your great work!
Since the UADA attack takes quite a long time, I was wondering if you have tried speeding it up by:
If so, did any of these work? Also, how sensitive was the performance to these settings?
Thank you in advance!
python roboticAttack/VLAAttacker/UADA_wrapper.py \ --maskidx 0 \ --lr 2e-3 \ --server $current_dir \ --device 0 \ --iter 2000 \ --accumulate 1 \ --bs 8 \ --warmup 20 \ --tags "debug testrun" \ --filterGripTrainTo1 false \ --geometry true \ --patch_size "3,50,50" \ --wandb_project "false" \ --wandb_project "false" \ --wandb_entity "xxx" \ --innerLoop 50 \ --dataset "libero_spatial" # "libero_spatial" / "libero_10" / "libero_goal" / "libero_goal" / "bridge_orig"