The Parallel Transfer of Task

With a distinction made between two forms of task knowledge transfer, representational and functional, MTL, a modi ed version of the MTL method of functional (parallel) transfer, is introduced. The MTL method employs a separate learning rate, k, for each task output node k. k varies as a function of a measure of relatedness, Rk, between the kth task and the primary task of interest. Results of experiments demonstrate the ability of MTL to dynamically select the most related source task(s) for the functional transfer of prior domain knowledge. The MTL method of learning is nearly equivalent to standard MTL when all parallel tasks are su ciently related to the primary task, and is similar to single task learning when none of the parallel tasks are related to the primary task. 2