Zobrazeno 1 - 4
of 4
pro vyhledávání: '"Yan, Bobby"'
The rapid growth in the size of deep learning models strains the capabilities of traditional dense computation paradigms. Leveraging sparse computation has become increasingly popular for training and deploying large-scale models, but existing deep l
Externí odkaz:
http://arxiv.org/abs/2405.16883
Autor:
Liang, Percy, Bommasani, Rishi, Lee, Tony, Tsipras, Dimitris, Soylu, Dilara, Yasunaga, Michihiro, Zhang, Yian, Narayanan, Deepak, Wu, Yuhuai, Kumar, Ananya, Newman, Benjamin, Yuan, Binhang, Yan, Bobby, Zhang, Ce, Cosgrove, Christian, Manning, Christopher D., Ré, Christopher, Acosta-Navas, Diana, Hudson, Drew A., Zelikman, Eric, Durmus, Esin, Ladhak, Faisal, Rong, Frieda, Ren, Hongyu, Yao, Huaxiu, Wang, Jue, Santhanam, Keshav, Orr, Laurel, Zheng, Lucia, Yuksekgonul, Mert, Suzgun, Mirac, Kim, Nathan, Guha, Neel, Chatterji, Niladri, Khattab, Omar, Henderson, Peter, Huang, Qian, Chi, Ryan, Xie, Sang Michael, Santurkar, Shibani, Ganguli, Surya, Hashimoto, Tatsunori, Icard, Thomas, Zhang, Tianyi, Chaudhary, Vishrav, Wang, William, Li, Xuechen, Mai, Yifan, Zhang, Yuhui, Koreeda, Yuta
Publikováno v:
Published in Transactions on Machine Learning Research (TMLR), 2023
Language models (LMs) are becoming the foundation for almost all major language technologies, but their capabilities, limitations, and risks are not well understood. We present Holistic Evaluation of Language Models (HELM) to improve the transparency
Externí odkaz:
http://arxiv.org/abs/2211.09110
Machine learning models are trained to minimize the mean loss for a single metric, and thus typically do not consider fairness and robustness. Neglecting such metrics in training can make these models prone to fairness violations when training data a
Externí odkaz:
http://arxiv.org/abs/2202.01719
Autor:
Garcia, Rolando, Liu, Eric, Sreekanti, Vikram, Yan, Bobby, Dandamudi, Anusha, Gonzalez, Joseph E., Hellerstein, Joseph M., Sen, Koushik
In modern Machine Learning, model training is an iterative, experimental process that can consume enormous computation resources and developer time. To aid in that process, experienced model developers log and visualize program variables during train
Externí odkaz:
http://arxiv.org/abs/2006.07357