Sim2real transfer is primarily concerned with transferring policies trained in simulation to potentially noisy real world environments. A common problem associated with sim2real transfer is estimating the real-world environmental parameters to ground the simulated environment to. Although existing methods such as Domain Randomisation (DR) can produce robust policies by sampling from a distribution of parameters during training, there is no established method for identifying the parameters of the corresponding distribution for a given real-world setting. In this work, we propose Uncertainty-aware policy search (UncAPS), where we use Universal Policy Network (UPN) to store simulation-trained task-specific policies across the full range of environmental parameters and then subsequently employ robust Bayesian optimisation to craft robust policies for the given environment by combining relevant UPN policies in a DR like fashion. Such policy-driven grounding is expected to be more efficient as it estimates only task-relevant sets of parameters. Further, we also account for the estimation uncertainties in the search process to produce policies that are robust against both aleatoric and epistemic uncertainties. We empirically evaluate our approach in a range of noisy, continuous control environments, and show its improved performance compared to competing baselines.
Joint universal lossy coding and identification of stationary mixing sources with general alphabets
Beckman Institute for Advanced Science and Technology, University of Illinois(伊利诺伊大学贝克曼先进科学技术研究所);Duke University(杜克大学)
我们考虑的问题,联合通用可变速率有损编码和识别的参数类的固定$β$混合源一般(波兰)字母。压缩性能是衡量的拉格朗日,而识别性能是衡量的变化之间的距离真实源和估计源。假设源以足够快的速率混合并满足一定的平滑性和Vapnik-Chervonenkis可学习性条件,则对于有界度量失真,存在用于联合有损压缩和识别的通用方案,其拉格朗日冗余收敛到零,如块长度$n$趋于无穷大时的$\sqrt{V_n \log n /n}$,其中$V_n$是由源的$n$维边缘分布定义的某类决策区域的Vapnik-Chervonenkis维数;此外,对于每个$n$,解码器可以识别活动源的$n$维边缘,直到在变距离中的半径为$O(\sqrt{V_n\log n/n})$的球,最终概率为1。结果补充了几个例子的参数源满足正则性条件。
Disentangling Policy from Offline Task Representation Learning via Adversarial Data Augmentation
National Key Laboratory for Novel Software Technology(计算机软件新技术国家重点实验室);School of Artificial Intelligence(人工智能学院);Nanjing University(南京大学);Polixir Technologies(铂乐奇科技)
Universal Pose Pretraining for Generalizable Vision-Language-Action Policies
Tencent Robotics X(腾讯机器人X);Futian Laboratory(福田实验室);The Hong Kong University of Science and Technology(香港科学与技术大学);Fudan University(复旦大学);Shanghai Innovation Institute(上海创新研究院)
Bound by semanticity: universal laws governing the generalization-identification tradeoff
Dipartimento di Scienze Matematiche, Politecnico di Torino(都灵理工大学数学科学系);CENTAI Institute(CENTAI研究院);Network Science Institute, Northeastern University(东北大学网络科学研究所);Institute for Experiential AI, Northeastern University(东北大学体验人工智能研究所);NP Lab, Network Science Institute, Northeastern University London(东北大学伦敦网络科学研究所NP实验室);Department of Psychology, Princeton University(普林斯顿大学心理学系);Program in Cognitive Science, Dartmouth College(达特茅斯学院认知科学项目);Department of Psychology, University of Michigan(密歇根大学心理学系);Microsoft Research(微软研究院);Princeton Neuroscience Institute(普林斯顿神经科学研究所);Department of Physics, Northeastern University(东北大学物理系)
Universal Bovine Identification via Depth Data and Deep Metric Learning
Bristol Veterinary School, University of Bristol(布里斯托尔大学布里斯托尔兽医学院);University of Bristol(布里斯托尔大学);Centre for Machine Vision, University of the West of England(西英格兰大学机器视觉中心);University of the West of England(西英格兰大学);School of Computer Science, University of Bristol(布里斯托尔大学计算机科学学院)
Thermal Tracks: A Gaussian process-based framework for universal melting curve analysis enabling unconstrained hit identification in thermal proteome profiling experiments
Department of Chemistry, School of Humanities and Sciences, Stanford University(化学系、人文与科学学院、斯坦福大学);Stanford Chem-H, Stanford University(斯坦福化学-人类健康系、斯坦福大学)