Research Results from Chiba University Update Knowledge of Robotics (Addressing data imbalance in Sim2Real: ImbalSim2Real scheme and its application in finger j oint stiffness self-sensing for soft robot-assisted rehabilitation)
Research Results from Chiba University Update Knowledge of Robotics (Addressing data imbalance in Sim2Real: ImbalSim2Real scheme and its application in finger j oint stiffness self-sensing for soft robot-assisted rehabilitation)
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – A new study on robotics is now availab le. According to news originating from Chiba, Japan, by NewsRx correspondents, r esearch stated, “The simulation-to-reality (sim2real) problem is a common issue when deploying simulation-trained models to real-world scenarios, especially giv en the extremely high imbalance between simulation and real-world data (scarce r eal-world data).” The news editors obtained a quote from the research from Chiba University: “Alth ough the cycleconsistent generative adversarial network (CycleGAN) has demonstr ated promise in addressing some sim2real issues, it encounters limitations in si tuations of data imbalance due to the lower capacity of the discriminator and th e indeterminacy of learned sim2real mapping. To overcome such problems, we propo sed the imbalanced Sim2Real scheme (ImbalSim2Real). Differing from CycleGAN, the ImbalSim2Real scheme segments the dataset into paired and unpaired data for two -fold training. The unpaired data incorporated discriminator-enhanced samples to further squash the solution space of the discriminator, for enhancing the discr iminator’s ability. For paired data, a term targeted regression loss was integra ted to ensure specific and quantitative mapping and further minimize the solutio n space of the generator. The ImbalSim2Real scheme was validated through numeric al experiments, demonstrating its superiority over conventional sim2real methods .”
NETL
NSTL