no code implementations • 22 Apr 2024 • Jiachen T. Wang, Zhun Deng, Hiroaki Chiba-Okabe, Boaz Barak, Weijie J. Su
Generative artificial intelligence (AI) systems are trained on large data corpora to generate new pieces of text, images, videos, and other media.
no code implementations • 8 Mar 2024 • Huiying Zhong, Zhun Deng, Weijie J. Su, Zhiwei Steven Wu, Linjun Zhang
Our work \textit{initiates} the theoretical study of multi-party RLHF that explicitly models the diverse preferences of multiple individuals.
no code implementations • 3 Jan 2024 • Haonan Wang, James Zou, Michael Mozer, Anirudh Goyal, Alex Lamb, Linjun Zhang, Weijie J Su, Zhun Deng, Michael Qizhe Xie, Hannah Brown, Kenji Kawaguchi
With the rise of advanced generative AI models capable of tasks once reserved for human creativity, the study of AI's creative potential becomes imperative for its responsible development and application.
no code implementations • 20 Dec 2023 • Jiachen Zhao, Zhun Deng, David Madras, James Zou, Mengye Ren
As the number of large language models (LLMs) released to the public grows, there is a pressing need to understand the safety implications associated with these models learning from third-party custom finetuning data.
1 code implementation • 22 Nov 2023 • Thomas P. Zollo, Todd Morrill, Zhun Deng, Jake C. Snell, Toniann Pitassi, Richard Zemel
The recent explosion in the capabilities of large language models has led to a wave of interest in how best to prompt a model to perform a given task.
1 code implementation • 1 Oct 2023 • Yiyang Zhou, Chenhang Cui, Jaehong Yoon, Linjun Zhang, Zhun Deng, Chelsea Finn, Mohit Bansal, Huaxiu Yao
Large vision-language models (LVLMs) have shown remarkable abilities in understanding visual information with human languages.
no code implementations • NeurIPS 2023 • Zhun Deng, Thomas P. Zollo, Jake C. Snell, Toniann Pitassi, Richard Zemel
Explicit finite-sample statistical guarantees on model performance are an important ingredient in responsible machine learning.
1 code implementation • 30 May 2023 • Kenji Kawaguchi, Zhun Deng, Xu Ji, Jiaoyang Huang
In this paper, we provide the first rigorous learning theory for justifying the benefit of information bottleneck in deep learning by mathematically relating information bottleneck to generalization errors.
2 code implementations • 8 Apr 2023 • Yuzhen Mao, Zhun Deng, Huaxiu Yao, Ting Ye, Kenji Kawaguchi, James Zou
As machine learning has been deployed ubiquitously across applications in modern data science, algorithmic fairness has become a great concern.
no code implementations • 8 Mar 2023 • Zhun Deng, Cynthia Dwork, Linjun Zhang
Fairness is captured by incorporating demographic subgroups into the class of functions~$\mathcal{C}$.
1 code implementation • 13 Feb 2023 • Ryumei Nakada, Halil Ibrahim Gulluk, Zhun Deng, Wenlong Ji, James Zou, Linjun Zhang
We show that the algorithm can detect the ground-truth pairs and improve performance by fully exploiting unpaired datasets.
1 code implementation • 27 Dec 2022 • Jake C. Snell, Thomas P. Zollo, Zhun Deng, Toniann Pitassi, Richard Zemel
In this work, we propose a flexible framework to produce a family of bounds on quantiles of the loss distribution incurred by a predictor.
no code implementations • 8 Nov 2022 • Zhun Deng, He Sun, Zhiwei Steven Wu, Linjun Zhang, David C. Parkes
AI methods are used in societally important settings, ranging from credit to employment to housing, and it is crucial to provide fairness in regard to algorithmic decision making.
1 code implementation • 7 Nov 2022 • Jiayao Zhang, Hongming Zhang, Zhun Deng, Dan Roth
We distill several insights from our analysis on study the peer review process with the help of large LMs.
no code implementations • 27 Jun 2022 • Kenji Kawaguchi, Zhun Deng, Kyle Luh, Jiaoyang Huang
This paper proves that robustness implies generalization via data-dependent generalization bounds.
no code implementations • 6 Jun 2022 • Zhun Deng, Jiayao Zhang, Linjun Zhang, Ting Ye, Yates Coley, Weijie J. Su, James Zou
Specifically, FIFA encourages both classification and fairness generalization and can be flexibly combined with many existing fair learning methods with logits-based losses.
no code implementations • 4 Nov 2021 • Maya Burhanpurkar, Zhun Deng, Cynthia Dwork, Linjun Zhang
Predictors map individual instances in a population to the interval $[0, 1]$.
no code implementations • 6 Oct 2021 • Wenlong Ji, Zhun Deng, Ryumei Nakada, James Zou, Linjun Zhang
Contrastive learning has achieved state-of-the-art performance in various self-supervised learning tasks and even outperforms its supervised counterpart.
no code implementations • ICLR 2022 • Wenlong Ji, Yiping Lu, Yiliang Zhang, Zhun Deng, Weijie J. Su
We prove that gradient flow on this model converges to critical points of a minimum-norm separation problem exhibiting neural collapse in its global minimizer.
no code implementations • 28 Jun 2021 • Kenji Kawaguchi, Linjun Zhang, Zhun Deng
Representation learning allows us to automatically discover suitable representations from raw sensory data.
no code implementations • NeurIPS 2021 • Zhun Deng, Linjun Zhang, Kailas Vodrahalli, Kenji Kawaguchi, James Zou
Recent works empirically demonstrate that adversarial training in the source data can improve the ability of models to transfer to new domains.
no code implementations • NeurIPS 2021 • Wenlong Ji, Yiping Lu, Yiliang Zhang, Zhun Deng, Weijie J Su
In this paper, we derive a landscape analysis to the surrogate model to study the inductive bias of the neural features and parameters from neural networks with cross-entropy.
no code implementations • 11 Feb 2021 • Linjun Zhang, Zhun Deng, Kenji Kawaguchi, James Zou
In addition, we study how Mixup improves calibration in semi-supervised learning.
no code implementations • 27 Oct 2020 • Zhun Deng, Hangfeng He, Weijie J. Su
Given that, we propose \emph{locally elastic stability} as a weaker and distribution-dependent stability notion, which still yields exponential generalization bounds.
no code implementations • ICML 2020 • Zhun Deng, Hangfeng He, Jiaoyang Huang, Weijie J. Su
An acknowledged weakness of neural networks is their vulnerability to adversarial perturbations to the inputs.
no code implementations • ICLR 2021 • Linjun Zhang, Zhun Deng, Kenji Kawaguchi, Amirata Ghorbani, James Zou
For robustness, we show that minimizing the Mixup loss corresponds to approximately minimizing an upper bound of the adversarial loss.
no code implementations • ICML 2020 • Zhun Deng, Cynthia Dwork, Jialiang Wang, Linjun Zhang
Robust optimization has been widely used in nowadays data science, especially in adversarial training.
no code implementations • 26 Sep 2020 • He Sun, Zhun Deng, Hui Chen, David C. Parkes
We introduce the decision-aware time-series conditional generative adversarial network (DAT-CGAN) as a method for time-series generation.
no code implementations • 20 Jun 2020 • Zhun Deng, Frances Ding, Cynthia Dwork, Rachel Hong, Giovanni Parmigiani, Prasad Patil, Pragya Sur
We study an adversarial loss function for $k$ domains and precisely characterize its limiting behavior as $k$ grows, formalizing and proving the intuition, backed by experiments, that observing data from a larger number of domains helps.
no code implementations • 15 Jun 2020 • Zhun Deng, Linjun Zhang, Amirata Ghorbani, James Zou
In this work, we investigate how adversarial robustness can be enhanced by leveraging out-of-domain unlabeled data.
no code implementations • 4 Jun 2019 • Zhun Deng, Cynthia Dwork, Jialiang Wang, Yao Zhao
We provide a general framework for characterizing the trade-off between accuracy and robustness in supervised learning.