Publikationen von L Schulze Buschoff
Alle Typen
Konferenzbeitrag (2)
1.
Konferenzbeitrag
The Acquisition of Physical Knowledge in Generative Neural Networks. In: 40th International Conference on Machine Learning, S. 30321 - 30341 (Hg. Krause, A.; Brunskill, E.; Cho, K.; Engelhardt, B.; Sabato, S. et al.). Fortieth International Conference on Machine Learning (ICML 2023), Honolulu, HI,USA, 23. Juli 2023 - 29. Juli 2023. (2023)
2.
Konferenzbeitrag
ImageNet suffers from dichotomous data difficulty. In: NeurIPS 2021 Workshop on ImageNet: past, present, and future, S. 1 - 27. NeurIPS 2021 Workshop on ImageNet: past, present, and future, 13. Dezember 2021 - 14. Dezember 2021. (2021)
Forschungspapier (1)
3.
Forschungspapier
Trivial or impossible: dichotomous data difficulty masks model differences (on ImageNet and beyond). (eingereicht)
Preprint (1)
4.
Preprint
Stochastic Gradient Descent Captures How Children Learn About Physics. (eingereicht)