Sökning: "Minnesförstärkta neurala nätverk"

Hittade 2 uppsatser innehållade orden Minnesförstärkta neurala nätverk.

  1. 1. Memory and Reasoning in Deep Learning : Data efficiency of the SAM-based Two-memory (STM) Model

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Andrzej Perzanowski; [2022]
    Nyckelord :Deep learning; Data efficiency; Memory-augmented neural networks; Memory; Reasoning; bAbI challenge; Djupinlärning; Dataeffektivitet; Minnesförstärkta nätverk; Minne; Resonemang; bAbI-utmaning;

    Sammanfattning : Developing Deep Learning models capable of learning to reason and store memories are some of the most important current challenges in AI research. Finding out which network architectures are best suited for tackling this problem can guide research toward the most promising approaches. LÄS MER

  3. 2. Improving Training of Differentiable Neural Computers on Time Series

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Isak Persson; [2022]
    Nyckelord :Memory augmented neural networks; Differentiable neural computers; Recurrent neural networks; Time series; Transfer learning; Minnesförstärkta neurala nätverk; Differentierbara neurala datorer; Återkommande neurala nätverk; Tidsserier; Överföra lärande;

    Sammanfattning : Memory Augmented Neural Networks (MANN) is a hot research area within deep learning. One of the most promising MANN is the Differentiable Neural Network (DNC) which is able to learn, in a fully differentiable way, how to represent and store data into an external memory. LÄS MER