Search Results - "Richter, Mats L"

  • Showing 1 - 14 results of 14
Refine Results
  1. 1
  2. 2

    AI-Based Crop Rotation for Sustainable Agriculture Worldwide by Schoning, Julius, Richter, Mats L.

    “…Artificial intelligence (AI) and sustainability. Two words not commonly used in the context of crop rotation management. However, simple AI-based expert…”
    Get full text
    Conference Proceeding
  3. 3

    Receptive Field Refinement for Convolutional Neural Networks Reliably Improves Predictive Performance by Richter, Mats L, Pal, Christopher

    Published 26-11-2022
    “…Minimal changes to neural architectures (e.g. changing a single hyperparameter in a key layer), can lead to significant gains in predictive performance in…”
    Get full text
    Journal Article
  4. 4

    CarbonSense: A Multimodal Dataset and Baseline for Carbon Flux Modelling by Fortier, Matthew, Richter, Mats L, Sonnentag, Oliver, Pal, Chris

    Published 07-06-2024
    “…Terrestrial carbon fluxes provide vital information about our biosphere's health and its capacity to absorb anthropogenic CO$_2$ emissions. The importance of…”
    Get full text
    Journal Article
  5. 5

    Should You Go Deeper? Optimizing Convolutional Neural Network Architectures without Training by Receptive Field Analysis by Richter, Mats L, Schöning, Julius, Wiedenroth, Anna, Krumnack, Ulf

    Published 05-10-2021
    “…When optimizing convolutional neural networks (CNN) for a specific image-based task, specialists commonly overshoot the number of convolutional layers in their…”
    Get full text
    Journal Article
  6. 6

    Wuerstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models by Pernias, Pablo, Rampas, Dominic, Richter, Mats L, Pal, Christopher J, Aubreville, Marc

    Published 01-06-2023
    “…The Twelfth International Conference on Learning Representations (ICLR), 2024 We introduce W\"urstchen, a novel architecture for text-to-image synthesis that…”
    Get full text
    Journal Article
  7. 7

    Simple and Scalable Strategies to Continually Pre-train Large Language Models by Ibrahim, Adam, Thérien, Benjamin, Gupta, Kshitij, Richter, Mats L, Anthony, Quentin, Lesort, Timothée, Belilovsky, Eugene, Rish, Irina

    Published 13-03-2024
    “…Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more…”
    Get full text
    Journal Article
  8. 8

    Should You Go Deeper? Optimizing Convolutional Neural Network Architectures without Training by Richter, Mats L., Schoning, Julius, Wiedenroth, Anna, Krumnack, Ulf

    “…When optimizing convolutional neural networks (CNN) for a specific image-based task, specialists commonly overshoot the number of convolutional layers in their…”
    Get full text
    Conference Proceeding
  9. 9

    Continual Pre-Training of Large Language Models: How to (re)warm your model? by Gupta, Kshitij, Thérien, Benjamin, Ibrahim, Adam, Richter, Mats L, Anthony, Quentin, Belilovsky, Eugene, Rish, Irina, Lesort, Timothée

    Published 07-08-2023
    “…Large language models (LLMs) are routinely pre-trained on billions of tokens, only to restart the process over again once new data becomes available. A much…”
    Get full text
    Journal Article
  10. 10

    Exploring the Properties and Evolution of Neural Network Eigenspaces during Training by Richter, Mats L, Malihi, Leila, Windler, Anne-Kathrin Patricia, Krumnack, Ulf

    Published 17-06-2021
    “…In this work we explore the information processing inside neural networks using logistic regression probes \cite{probes} and the saturation metric…”
    Get full text
    Journal Article
  11. 11

    Spectral Analysis of Latent Representations by Shenk, Justin, Richter, Mats L, Arpteg, Anders, Huss, Mikael

    Published 19-07-2019
    “…We propose a metric, Layer Saturation, defined as the proportion of the number of eigenvalues needed to explain 99% of the variance of the latent…”
    Get full text
    Journal Article
  12. 12

    Size Matters by Richter, Mats L, Byttner, Wolf, Krumnack, Ulf, Schallner, Ludwdig, Shenk, Justin

    Published 09-02-2021
    “…Artificial Neural Networks and Machine Learning ICANN 2021 133-144 Fully convolutional neural networks can process input of arbitrary size by applying a…”
    Get full text
    Journal Article
  13. 13

    Feature Space Saturation during Training by Richter, Mats L, Shenk, Justin, Byttner, Wolf, Arpteg, Anders, Huss, Mikael

    Published 15-06-2020
    “…British Machine Vision Conference (BMVC) 2021 We propose layer saturation - a simple, online-computable method for analyzing the information processing in…”
    Get full text
    Journal Article
  14. 14