Timezone: »

 
Poster
Importance-aware Co-teaching for Offline Model-based Optimization
Ye Yuan · Can Chen · Zixuan Liu · Willie Neiswanger · Xue (Steve) Liu

Wed Dec 13 08:45 AM -- 10:45 AM (PST) @ Great Hall & Hall B1+B2 #304
in Poster Session 3 »
Offline model-based optimization aims to find a design that maximizes a property of interest using only an offline dataset, with applications in robot, protein, and molecule design, among others. A prevalent approach is gradient ascent, where a proxy model is trained on the offline dataset and then used to optimize the design. This method suffers from an out-of-distribution issue, where the proxy is not accurate for unseen designs. To mitigate this issue, we explore using a pseudo-labeler to generate valuable data for fine-tuning the proxy. Specifically, we propose $\textit{\textbf{I}mportance-aware \textbf{C}o-\textbf{T}eaching for Offline Model-based Optimization}~(\textbf{ICT})$. This method maintains three symmetric proxies with their mean ensemble as the final proxy, and comprises two steps. The first step is $\textit{pseudo-label-driven co-teaching}$. In this step, one proxy is iteratively selected as the pseudo-labeler for designs near the current optimization point, generating pseudo-labeled data. Subsequently, a co-teaching process identifies small-loss samples as valuable data and exchanges them between the other two proxies for fine-tuning, promoting knowledge transfer. This procedure is repeated three times, with a different proxy chosen as the pseudo-labeler each time, ultimately enhancing the ensemble performance.To further improve accuracy of pseudo-labels, we perform a secondary step of $\textit{meta-learning-based sample reweighting}$,which assigns importance weights to samples in the pseudo-labeled dataset and updates them via meta-learning. ICT achieves state-of-the-art results across multiple design-bench tasks, achieving the best mean rank $3.1$ and median rank $2$ among $15$ methods.Our source code can be accessed here.

Author Information

Ye Yuan (McGill University)
Ye Yuan

Hello, This is Ye (Steven) Yuan, a first-year Ph.D. student from McGill University's CPS Lab. I am fortunate to be supervised by Professor Xue (Steve) Liu. I also obtained my Bachelor of Science degree in Honours Computer Science from McGill University as well. My research interests are in generative models and meta-learning. Please feel free to contact me if you are interested in collaborating with me.

Can Chen (Mila - Quebec AI Institute)
Zixuan Liu (University of Washington)
Willie Neiswanger (Stanford / USC)
Xue (Steve) Liu (McGill University)

More from the Same Authors