Neural networks in the automotive sector commonly have to process varying number of objects per observation. Deep Set feature extractors have shown great success on problems in reinforcement learning with dynamic observations, achieving state-of-the-art performance on common tasks like highway driving. However, recent work has shown that neural networks suffer from a spectral bias, fitting to low frequencies of scalar features used in such representations, like velocities and distances. We introduce a novel set feature extractor combining learned Fourier features with the Deep Set architecture. The proposed architecture reduces this spectral bias allowing more sample-efficient training and better performance by learning a more detailed representation. Extensive experiments are conducted on three different environments, including two novel environments featuring a large number of objects in challenging random scenarios. Our method outperforms state-of-the-art approaches and exceeds the performance of existing Deep Set architectures on these challenging new tasks.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Learned Fourier Bases for Deep Set Feature Extractors in Automotive Reinforcement Learning


    Contributors:


    Publication date :

    2023-09-24


    Size :

    1157779 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    WATER EXTRACTORS AND METHODS OF MAKING WATER EXTRACTORS

    PALMER NIGEL | European Patent Office | 2020

    Free access

    Water extractors and methods of making water extractors

    PALMER NIGEL G M | European Patent Office | 2021

    Free access

    WATER EXTRACTORS AND METHODS OF MAKING WATER EXTRACTORS

    PALMER NIGEL G M | European Patent Office | 2021

    Free access

    Feature Extraction and Image Registration of Color Images using Fourier Bases

    Jagannadan, V. / Prakash, M. C. / Sarma, R. R. et al. | British Library Conference Proceedings | 2005


    Convolutional Neural Networks as Classification Tools and Feature Extractors for Distinguishing Malware Programs

    Priyamvada Davuluru, Venkata Salini / Narayanan Narayanan, Barath / Balster, Eric J. | IEEE | 2019