Why the validation loss is smaller than the training loss?

[benjamin3344]

I have replicated the NSVF datasets. The rendered views looked good. But why the training loss is always far larger than the validation loss? Here are the scripts of parameters.

python -u train.py ${Dataset}
--user-dir fairnr
--task single_object_rendering
--train-views "0..100" --view-resolution "800x800"
--max-sentences 1 --view-per-batch 1 --pixel-per-view 512
--no-preload
--sampling-on-mask 1.0 --no-sampling-at-reader
--valid-views "100..200" --valid-view-resolution "800x800"
--valid-view-per-batch 1
--transparent-background "1.0,1.0,1.0" --background-stop-gradient
--arch nsvf_base
--initial-boundingbox ${SOURCEDIR}/nsvf/Spaceship/bbox.txt
--use-octree
--raymarching-stepsize-ratio 0.125
--discrete-regularization
--color-weight 128.0 --alpha-weight 1.0
--optimizer "adam" --adam-betas "(0.9, 0.999)"
--lr 0.001 --lr-scheduler "polynomial_decay" --total-num-update 150000
--criterion "srn_loss" --clip-norm 0.0
--num-workers 0
--seed 2
--save-interval-updates 500 --max-update 150000
--virtual-epoch-steps 5000 --save-interval 1
--half-voxel-size-at "5000,25000,75000"
--reduce-step-size-at "5000,25000,75000"
--pruning-every-steps 2500
--keep-interval-updates 5 --keep-last-epochs 5
--log-format simple --log-interval 1
--save-dir ${Result}/${MODEL}
--tensorboard-logdir ${Result}/tensorboard/${MODEL}
| tee -a ${Result}/train.log