Tag
The author shares insights on building AI agents for small businesses, emphasizing that the hardest part is selecting the right task to automate, not the agent itself. Key considerations include volume, input structure, cost of errors, and existence of manual processes.
This paper introduces a framework for when transfer should be expected in continual learning and proposes Transfer-Selective Replay (TSR), which selects replay data predicted to benefit the incoming task rather than indiscriminately replaying past examples. TSR improves forward transfer while maintaining stability, outperforming existing replay baselines.