Skip to main content

Table 6 Tests on two independent experiments.

From: Multi-task learning for cross-platform siRNA efficacy prediction: an in-silico study

Tests RMSE
   IE1 IE2
Test 4 (50% training data) Single task learning 34.1116 35.8600
  Multi-task learning 29.7394 30.5459
Test 5 (with added tasks, 50% training data) 26.6910 26.1009
  1. "E" denotes "Experiment". Test 4: Comparison between single task learning and multi-task learning on the two independent experiments, both trained with 50% of the data from each experiment, respectively. Test 5: Multi-task learning on the two independent experiments together with the former 14 experiments, totally 16 experiments, trained with 50% of the data from each experiment, respectively.