Multi-agent active perception with prediction rewards

Open in new window