Zobrazeno 1 - 10
of 260
pro vyhledávání: '"Dhillon, Inderjit S"'
Data pruning, the combinatorial task of selecting a small and informative subset from a large dataset, is crucial for mitigating the enormous computational costs associated with training data-hungry modern deep learning models at scale. Since large-s
Externí odkaz:
http://arxiv.org/abs/2406.17188
Autor:
Das, Rudrajit, Dhillon, Inderjit S., Epasto, Alessandro, Javanmard, Adel, Mao, Jieming, Mirrokni, Vahab, Sanghavi, Sujay, Zhong, Peilin
The performance of a model trained with \textit{noisy labels} is often improved by simply \textit{retraining} the model with its own predicted \textit{hard} labels (i.e., $1$/$0$ labels). Yet, a detailed theoretical characterization of this phenomeno
Externí odkaz:
http://arxiv.org/abs/2406.11206
There is a notable dearth of results characterizing the preconditioning effect of Adam and showing how it may alleviate the curse of ill-conditioning -- an issue plaguing gradient descent (GD). In this work, we perform a detailed analysis of Adam's p
Externí odkaz:
http://arxiv.org/abs/2402.07114
Large language models (LLMs) have demonstrated remarkable capabilities in solving complex open-domain tasks, guided by comprehensive instructions and demonstrations provided in the form of prompts. However, these prompts can be lengthy, often compris
Externí odkaz:
http://arxiv.org/abs/2311.10117
Autor:
Wang, Yihan, Si, Si, Li, Daliang, Lukasik, Michal, Yu, Felix, Hsieh, Cho-Jui, Dhillon, Inderjit S, Kumar, Sanjiv
Pretrained large language models (LLMs) are general purpose problem solvers applicable to a diverse set of tasks with prompts. They can be further improved towards a specific task by fine-tuning on a specialized dataset. However, fine-tuning usually
Externí odkaz:
http://arxiv.org/abs/2211.00635
Extreme multi-label classification (XMC) is a popular framework for solving many real-world problems that require accurate prediction from a very large number of potential output choices. A popular approach for dealing with the large label space is t
Externí odkaz:
http://arxiv.org/abs/2210.08410
Approximate K-Nearest Neighbor Search (AKNNS) has now become ubiquitous in modern applications, for example, as a fast search procedure with two tower deep learning models. Graph-based methods for AKNNS in particular have received great attention due
Externí odkaz:
http://arxiv.org/abs/2206.11408
Conventional methods for query autocompletion aim to predict which completed query a user will select from a list. A shortcoming of this approach is that users often do not know which query will provide the best retrieval performance on the current i
Externí odkaz:
http://arxiv.org/abs/2204.10936
Data augmentation is popular in the training of large neural networks; currently, however, there is no clear theoretical comparison between different algorithmic choices on how to use augmented data. In this paper, we take a step in this direction -
Externí odkaz:
http://arxiv.org/abs/2202.12230
Autor:
Chien, Eli, Chang, Wei-Cheng, Hsieh, Cho-Jui, Yu, Hsiang-Fu, Zhang, Jiong, Milenkovic, Olgica, Dhillon, Inderjit S
Learning on graphs has attracted significant attention in the learning community due to numerous real-world applications. In particular, graph neural networks (GNNs), which take numerical node features and graph structure as inputs, have been shown t
Externí odkaz:
http://arxiv.org/abs/2111.00064