Research map / Reinforcement learning / Imitation learning
Variational Adversarial Kernel Learned Imitation Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| learned-kernel MMD with variational bottleneck instead of fixed-kernel MMD (GMMIL) | GMMIL · Walker2D | reward of learned policy: 706.0 -> 735.2 | Helped |
| learned-kernel MMD instead of fixed kernel | GMMIL · HumanoidFlagrun | reward of learned policy: 111.9 -> 262.8 | Helped |
| MMD-VKL cost instead of JS divergence (GAIL) | GAIL · Ant | reward of learned policy: 1780.0 -> 2348.8 | Helped |
| MMD-VKL instead of JS divergence with information bottleneck (VAIL) | VAIL · Walker2D | reward of learned policy: 257.1 -> 735.2 | Helped |
| random Fourier kernel (learn base kernel) instead of Gaussian kernel (ablation) | VAKLIL-G · Humanoid | reward of learned policy: 188.2 -> 198.7 | Helped |
| random Fourier kernel instead of Gaussian kernel (ablation) | VAKLIL-G · HumanoidFlagrun | reward of learned policy: 217.6 -> 262.8 | Helped |
| VAKLIL-F instead of GAIL on traffic control | GAIL · Traffic control TRPE | TRPE: 56.70 -> 79.45 | Helped |
| VAKLIL-F instead of GAIL on traffic control | GAIL · Traffic control VOR | vehicles on road (VOR, lower better): 0.49 -> 1.15 | Hurt |
Closest papers by meaning
- Imitation Learning via Kernel Mean Embedding (AAAI 2018 · similarity 0.93)
- Visual Adversarial Imitation Learning using Variational Models (NeurIPS 2021 · similarity 0.90)
- Learning from Demonstration: Provably Efficient Adversarial Policy Imitation with Linear Function Approximation (ICML 2022 · similarity 0.89)
- Deterministic and Discriminative Imitation (D2-Imitation): Revisiting Adversarial Imitation for Sample Efficiency (AAAI 2022 · similarity 0.89)
- Latent Wasserstein Adversarial Imitation Learning (ICLR 2026 · similarity 0.88)
- Adversarial Imitation Learning via Boosting (ICLR 2024 · similarity 0.87)
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.