Success Conditioning as Policy Improvement: The Optimization Problem Solved by Imitating Success

Open in new window