AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Myriad: a real-world testbed to bridge trajectory optimization and deep learning

Nikolaus H. R. Howe, Simon Dufort-Labbé, Nitarshan Rajkumar, Pierre-Luc Bacon

NeurIPS 2022 · Reinforcement learning › Visual and representation RL · approach: RL for combinatorial routing · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train end-to-end through unrolled NLP solver instead of sysid then planning on Cart-PoleNeural ODE model · Cart-Pole Swing-Upcost: 242.25 -> 36.90Improved
end-to-end training instead of Neural ODE sysid on PendulumNeural ODE model · Pendulumcost: 20.88 -> 1.90Improved
end-to-end training instead of Neural ODE sysid on Predator PreyNeural ODE model · Predator Preycost: 1.95 -> 1.04Improved
end-to-end training instead of Neural ODE sysid on Van Der PolNeural ODE model · Van Der Polcost: 11.12 -> 16.81Hurt
end-to-end training instead of Neural ODE sysid on Mountain CarNeural ODE model · Mountain Carcost: 15.87 -> 3000Hurt
end-to-end training instead of Neural ODE sysid on Timber HarvestNeural ODE model · Timber Harvestcost: -3928.77 -> -500.00Hurt
end-to-end training instead of Neural ODE sysid on BacteriaNeural ODE model · Bacteriacost: -2.72 -> -7.60Improved
end-to-end training instead of Neural ODE sysid on TumourNeural ODE model · Tumourcost: 8468.68 -> 9161.23Hurt
learn structured parametric model instead of using true parameterstrajectory optimization · Pendulumcost: 25.75 -> 25.49No clear effect
learn Neural ODE instead of parametric model for dynamicstrajectory optimization · Cart-Pole Swing-Upcost: 87.78 -> 242.25Hurt
smaller training set of 10 trajectories instead of 100 for parametric sysid (ablation)parametric sysid · parametric environmentscostNo clear effect
use 5 implicit planning steps per iteration (ablation)end-to-end algorithm · end-to-endcostImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy