publications

publications by categories in reversed chronological order. generated by jekyll-scholar.

2026

  1. Mitigating Reward Hacking in RLHF via Advantage Sign Robustness
    Shinnosuke Ono, Johannes Ackermann, Soichiro Nishimori, Takashi Ishida, and Masashi Sugiyama
    Jul 2026
    arXiv
  2. ICML
    gradregrl.png
    Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards
    Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, and Masashi Sugiyama
    In ICML 2026 , Jul 2026
    arXiv PDF Website
  3. ICML
    bridgingBBO.png
    Bridging Spherical Black-Box Optimizers
    Johannes Ackermann, and Stefano Peluchetti
    In ICML 2026 , Jul 2026

2025

  1. COLM
    ocrm.png
    Off-Policy Corrected Reward Modeling for Reinforcement Learning from Human Feedback
    Johannes Ackermann, Takashi Ishida, and Masashi Sugiyama
    In Conference on Language Modeling (COLM) 2025 , Oct 2025
    PDF
  2. RLC
    recursive_reward_aggregation.png
    Recursive Reward Aggregation
    Yuting Tang, Yivan Zhang, Johannes Ackermann, Yu-Jie Zhang, Soichiro Nishimori, and Masashi Sugiyama
    In Reinforcement Learning Conference (RLC) 2025 , Aug 2025
    PDF
  3. RLC
    pu_offlinerl.png
    Offline Reinforcement Learning with Domain-Unlabeled Data
    Soichiro Nishimori, Xin-Qiang Cai, Johannes Ackermann, and Masashi Sugiyama
    In Reinforcement Learning Conference (RLC) 2025 , Aug 2025
    arXiv PDF

2024

  1. RLC
    nonstationary_dataset.png
    Offline Reinforcement Learning from Datasets with Structured Non-Stationarity
    Johannes Ackermann, Takayuki Osa, and Masashi Sugiyama
    In Reinforcement Learning Conference (RLC) 2024 , Aug 2024
    arXiv PDF Website

2022

  1. Workshop
    msbd_overview.jpg
    High-Resolution Image Editing via Multi-Stage Blended Diffusion
    Johannes Ackermann, and Minjun Li
    In NeurIPS Machine Learning for Creativity and Design Workshop 2022 , Dec 2022
    arXiv PDF Blog

2021

  1. ECML-PKDD
    task_clustering.png
    Unsupervised Task Clustering for Multi-Task Reinforcement Learning
    Johannes Ackermann, Oliver Richter, and Roger Wattenhofer
    In ECML-PKDD 2021 , Sep 2021
    PDF Video

2020

  1. ECOC
    qam_image.png
    Convolutional Neural Network Based Blind Estimation of Generalized Mutual Information for Optical Communication
    Johannes Ackermann, Maximilian Schädler, and Christian Blümm
    In European Conference on Optical Communication (ECOC) , Dec 2020
    PDF

2019

  1. Workshop
    estimation_matd3.png
    Reducing Overestimation Bias in Multi-Agent Domains Using Double Centralized Critics
    Johannes Ackermann, Volker Gabler, Takayuki Osa, and Masashi Sugiyama
    In Deep Reinforcement Learning Workshop at NeurIPS , Dec 2019
    arXiv PDF