[WIP] Implement DPPO algorithm on OGbench

[Copilot]

Thanks for asking me to work on this. I will get started on it and keep this PR's description up to date as I form a plan and make progress.

Original prompt

help me implement dppo algorthm on OGbench

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[Whiterrrrr]

@copilot Help me implement dppo (Diffusion Proximal Policy Optimization) algorithm on OGbench, buiding upon FQL (flow Q-learning) codebase

[Copilot]

@copilot Help me implement dppo (Diffusion Proximal Policy Optimization) algorithm on OGbench, buiding upon FQL (flow Q-learning) codebase

I'd like to clarify the requirements for implementing DPPO (Diffusion Proximal Policy Optimization):

1. I don't see any FQL (Flow Q-learning) codebase in this repository. Should I:

   • Import/integrate FQL from an external repository?
   • Build DPPO from scratch using the existing diffusion infrastructure (from BREEZE)?

2. I don't see OGbench integrated in the current codebase. The repository currently uses:

   • DMC (DeepMind Control) domains (walker, quadruped, jaco)
   • ExORL datasets

Should I:

• Add OGbench environment support?
• Adapt DPPO to work with the existing DMC/ExORL setup?

3. Could you provide:
   • Link to the FQL codebase you'd like me to build upon?
   • Link to OGbench repository/documentation?
   • Any specific paper references for DPPO?

Once I understand these requirements, I can create a proper implementation plan.