SANE2023 | Kyunghyun Cho - Beyond Test Accuracies for Studying Deep Neural Networks скачать в хорошем качестве

SANE2023 | Kyunghyun Cho - Beyond Test Accuracies for Studying Deep Neural Networks

Kyunghyun Cho, associate professor of computer science and data science at New York University and senior director of frontier research at Prescient Design, presents his work on going beyond test accuracies for studying deep neural networks at the SANE 2023 workshop at New York University, New York, on October 26, 2023. More info on the SANE workshop series: http://www.saneworkshop.org/ Abstract: Already in 2015, Leon Bottou discussed the prevalence and end of the training/test experimental paradigm in machine learning. The machine learning community has however continued to stick to this paradigm until now (2023), relying almost entirely and exclusively on the test-set accuracy, which is a rough proxy to the true quality of a machine learning system we want to measure. There are however many aspects in building a machine learning system that require more attention. Specifically, I will discuss three such aspects in this talk; (1) model assumption and construction, (2) optimization and (3) inference. For model assumption and construction, I will discuss our recent work on generative multitask learning and incidental correlation in multimodal learning. For optimization, I will talk about how we can systematically study and investigate learning trajectories. Finally for inference, I will lay out two consistencies that must be satisfied by a large-scale language model and demonstrate that most of the language models do not fully satisfy such consistencies.