Zobrazeno 1 - 10
of 453
pro vyhledávání: '"Saraò, A."'
Artificial neural networks often struggle with catastrophic forgetting when learning multiple tasks sequentially, as training on new tasks degrades the performance on previously learned ones. Recent theoretical work has addressed this issue by analys
Externí odkaz:
http://arxiv.org/abs/2409.18061
A wide range of empirical and theoretical works have shown that overparameterisation can amplify the performance of neural networks. According to the lottery ticket hypothesis, overparameterised networks have an increased chance of containing a sub-n
Externí odkaz:
http://arxiv.org/abs/2406.01589
Machine learning systems often acquire biases by leveraging undesired features in the data, impacting accuracy variably across different sub-populations. Current understanding of bias formation mostly focuses on the initial and final stages of learni
Externí odkaz:
http://arxiv.org/abs/2405.18296
Diverse studies in systems neuroscience begin with extended periods of curriculum training known as `shaping' procedures. These involve progressively studying component parts of more complex tasks, and can make the difference between learning a task
Externí odkaz:
http://arxiv.org/abs/2402.18361
Reinforcement learning (RL) algorithms have proven transformative in a range of domains. To tackle real-world domains, these systems often use neural networks to learn policies directly from pixels or other high-dimensional sensory input. By contrast
Externí odkaz:
http://arxiv.org/abs/2306.10404
Transfer learning is a powerful tool enabling model training with limited amounts of data. This technique is particularly useful in real-world problems where data availability is often a serious limitation. The simplest transfer learning protocol is
Externí odkaz:
http://arxiv.org/abs/2303.01429
Machine learning (ML) may be oblivious to human bias but it is not immune to its perpetuation. Marginalisation and iniquitous group representation are often traceable in the very data used for training, and may be reflected or even enhanced by the le
Externí odkaz:
http://arxiv.org/abs/2205.15935
Publikováno v:
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:12455-12477 (2022)
Continual learning - learning new tasks in sequence while maintaining performance on old tasks - remains particularly challenging for artificial neural networks. Surprisingly, the amount of forgetting does not increase with the dissimilarity between
Externí odkaz:
http://arxiv.org/abs/2205.09029
Autor:
Aiello, Ivano W., Ash, Jeanine L., Bojanova, Diana P., Buatier, Martine D., Edgcomb, Virginia P., Galerne, Christophe Y., Gontharet, Swanne, Heuer, Verena B., Jiang, Shijun, Kars, Myriam A.C., Koornneef, Louise M.T., Marsaglia, Kathleen M., Meyer, Nicolette R., Morono, Yuki, Negrete-Aranda, Raquel, Neumann, Florian, Peña-Salinas, Manet E., Pérez-Cruz, Ligia L., Ran, Lihua, Riboulleau, Armelle, Sarao, John A., Schubert, Florian, Singh, S. Khogenkumar, Stock, Joann M., Toffin, Laurent, Xie, Wei, Zhuang, Guangchao, Yamanaka, Toshiro, Sakamoto, Arisa, Kiyokawa, Kanon, Jo, Jaeguk, Onishi, Yuji, Kuwahara, Yoshihiro, Kim, Ji-Hoon, Pastor, Lucie C., Teske, Andreas, Lizarralde, Daniel, Höfig, Tobias W.
Publikováno v:
In Chemical Geology 20 October 2024 666
Autor:
Aiello, Ivano W., Höfig, Tobias W., Riboulleau, Armelle, Teske, Andreas P., Lizarralde, Daniel, Ash, Jeanine L., Bojanova, Diana P., Buatier, Martine D., Edgcomb, Virginia P., Galerne, Christophe Y., Gontharet, Swanne, Heuer, Verena B., Jiang, Shijun, Kars, Myriam A.C., Kim, Ji-Hoon, Koornneef, Louise M.T., Marsaglia, Kathleen M., Meyer, Nicolette R., Morono, Yuki, Negrete-Aranda, Raquel, Neumann, Florian, Pastor, Lucie C., Peña-Salinas, Manet E., Pérez-Cruz, Ligia L., Ran, Lihua, Sarao, John A., Schubert, Florian, Khogenkumar Singh, S., Stock, Joann M., Toffin, Laurent, Xie, Wei, Yamanaka, Toshiro, Zhuang, Guangchao
Publikováno v:
In Geochimica et Cosmochimica Acta 1 September 2024 380:71-82