Deutsch Intern
    Data Science Chair

    Parameter Efficient Neural Networks

    10/13/2023

    As neural networks grow larger and more complex to solve new challenging tasks, they are often no longer able to run on end-user hardware and incur significant power and storage costs. For this reason, there is a need for parameter-efficient solutions that deliver high performance while being resource-efficient. Thus, parameter-efficient neural networks are of high interest as they are crucial for enabling a broad deployment of artificial intelligence in different domains, and for keeping neural networks accessible, efficient, and environmentally friendly. This task can be approached from different perspectives, such as building models that adapt their size to current needs, or through smart parameter reuse. The specific topic can be chosen together with the supervisor.

    As neural networks grow larger and more complex to solve new challenging tasks, they are often no longer able to run on end-user hardware and incur significant power and storage costs. For this reason, there is a need for parameter-efficient solutions that deliver high performance while being resource-efficient. Thus, parameter-efficient neural networks are of high interest as they are crucial for enabling a broad deployment of artificial intelligence in different domains, and for keeping neural networks accessible, efficient, and environmentally friendly

    This task can be approached from different perspectives, such as building models that adapt their size to current needs, or through smart parameter reuse. The specific topic can be chosen together with the supervisor.

    Supervisor: Janna Omeliyanenko

    Back