6 papers
HALO: Hindsight-Augmented Learning for Online Auto-Bidding
Pusen Dong, Chenglong Cao, Xinyu Zhou +4
Digital advertising platforms operate millisecond-level auctions through Real-Time Bidding (RTB) systems, where advertisers compete for ad impressions through algorithmic bids. Thi…
HyperINF: Unleashing the HyperPower of the Schulz's Method for Data Influence Estimation
Xinyu Zhou, Simin Fan, Martin Jaggi
Influence functions provide a principled method to assess the contribution of individual training samples to a specific target. Yet, their high computational costs limit their appl…
Private Model Personalization Revisited
Conor Snedeker, Xinyu Zhou, Raef Bassily
We study model personalization under user-level differential privacy (DP) in the shared representation framework. In this problem, there are users whose data is statistically h…
NeuralGrok: Accelerate Grokking by Neural Gradient Transformation
Xinyu Zhou, Simin Fan, Martin Jaggi +1
Grokking is proposed and widely studied as an intricate phenomenon in which generalization is achieved after a long-lasting period of overfitting. In this work, we propose NeuralGr…
XNN: Paradigm Shift in Mitigating Identity Leakage within Cloud-Enabled Deep Learning
Kaixin Liu, Huixin Xiong, Bingyu Duan +4
In the domain of cloud-based deep learning, the imperative for external computational resources coexists with acute privacy concerns, particularly identity leakage. To address this…
LoGAH: Predicting 774-Million-Parameter Transformers using Graph HyperNetworks with 1/100 Parameters
Xinyu Zhou, Boris Knyazev, Alexia Jolicoeur-Martineau +1
A good initialization of deep learning models is essential since it can help them converge better and faster. However, pretraining large models is unaffordable for many researchers…