Transcription of Rainbow: Combining Improvements in Deep Reinforcement …
{{id}} {{{paragraph}}}
Example: marketing
Rainbow: Combining Improvements in Deep Reinforcement …
rainbow : Combining Improvements in deep Reinforcement LearningMatteo HesselDeepMindJoseph ModayilDeepMindHado van HasseltDeepMindTom SchaulDeepMindGeorg OstrovskiDeepMindWill DabneyDeepMindDan HorganDeepMindBilal PiotDeepMindMohammad AzarDeepMindDavid SilverDeepMindAbstractThe deep Reinforcement learning community has made sev-eral independent Improvements to the DQN algorithm. How-ever, it is unclear which of these extensions are complemen-tary and can be fruitfully combined. This paper examinessix extensions to the DQN algorithm and empirically studiestheir combination. Our experiments show that the combina-tion provides state-of-the-art performance on the Atari 2600benchmark, both in terms of data efficiency and final perfor-mance. We also provide results from a detailed ablation studythat shows the contribution of each component to overall many recent successes in scaling Reinforcement learn-ing (RL) to complex sequential decision-making problemswere kick-started by the deep Q-Networks algorithm (DQN;Mnih et al.)
Deep reinforcement learning and DQN. Large state and/or action spaces make it intractable to learn Q value estimates for each state and action pair independently. In deep reinforcement learning, we represent the various com-ponents of agents, such as policies ˇ(s;a) or values q(s;a), with deep (i.e., multi-layer) neural networks. The parameters
Loading..
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document: