This PR contains a number of fixes. These fixes make the peg insertion task succeed (at least for me) on all four conditions. The fixes fall into a few categories:
Changes to peg task to match the MATLAB reference:
changed hole positions (still 0.2 apart, but center in a different place) -- no idea if this makes a difference
changed cost function to raise the action cost and change wp
Changes to algorithm parameters
lowered dual_rate to 0.1 to match matlab reference
changed max samples for GMM prior to match matlab reference
raised # of SGD iterations (this seems to have made a difference)
Changes to default algorithm settings
lowered default polwt to 0.1 (1.0 is way too high for a default)
changed the polwt lowering/raising heuristic to match matlab reference
Deep copy bugfix
added deep copy when flipping prev/cur variables for iteration, this appears to make a difference
This PR contains a number of fixes. These fixes make the peg insertion task succeed (at least for me) on all four conditions. The fixes fall into a few categories: