Timezone: »
While unbiased machine learning models are essential for many applications, bias is a human-defined concept that can vary across tasks. Given only input-label pairs, algorithms may lack sufficient information to distinguish stable (causal) features from unstable (spurious) features. However, related tasks often share similar biases -- an observation we may leverage to develop stable classifiers in the transfer setting. In this work, we explicitly inform the target classifier about unstable features in the source tasks. Specifically, we derive a representation that encodes the unstable features by contrasting different data environments in the source task. We achieve robustness by clustering data of the target task according to this representation and minimizing the worst-case risk across these clusters. We evaluate our method on both text and image classifications. Empirical results demonstrate that our algorithm is able to maintain robustness on the target task for both synthetically generated environments and real-world environments. Our code is available at https://github.com/YujiaBao/Tofu.
Author Information
Yujia Bao (MIT)
Shiyu Chang (UCSB)
Regina Barzilay (MIT CSAIL)

Regina Barzilay is an Israeli-American computer scientist. She is a professor at the Massachusetts Institute of Technology and a faculty lead for artificial intelligence at the MIT Jameel Clinic. Her research interests are in natural language processing and applications of deep learning to chemistry and oncology.
Related Events (a corresponding poster, oral, or spotlight)
-
2022 Poster: Learning Stable Classifiers by Transferring Unstable Features »
Thu. Jul 21st through Fri the 22nd Room Hall E #507
More from the Same Authors
-
2023 : Optimizing protein fitness using Bi-level Gibbs sampling with Graph-based Smoothing »
Andrew Kirjner · Jason Yim · Raman Samusevich · Tommi Jaakkola · Regina Barzilay · Ila R. Fiete -
2023 : Optimizing protein fitness using Gibbs sampling with Graph-based Smoothing »
Andrew Kirjner · Jason Yim · Raman Samusevich · Tommi Jaakkola · Regina Barzilay · Ila R. Fiete -
2023 Poster: Towards Coherent Image Inpainting Using Denoising Diffusion Implicit Models »
Guanhua Zhang · Jiabao Ji · Yang Zhang · Mo Yu · Tommi Jaakkola · Shiyu Chang -
2023 Poster: PromptBoosting: Black-Box Text Classification with Ten Forward Passes »
Bairu Hou · Joe O'Connor · Jacob Andreas · Shiyu Chang · Yang Zhang -
2023 Poster: SE(3) diffusion model with application to protein backbone generation »
Jason Yim · Brian Trippe · Valentin De Bortoli · Emile Mathieu · Arnaud Doucet · Regina Barzilay · Tommi Jaakkola -
2022 Poster: Data-Efficient Double-Win Lottery Tickets from Robust Pre-training »
Tianlong Chen · Zhenyu Zhang · Sijia Liu · Yang Zhang · Shiyu Chang · Zhangyang “Atlas” Wang -
2022 Poster: Antibody-Antigen Docking and Design via Hierarchical Structure Refinement »
Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2022 Poster: Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness »
Tianlong Chen · Huan Zhang · Zhenyu Zhang · Shiyu Chang · Sijia Liu · Pin-Yu Chen · Zhangyang “Atlas” Wang -
2022 Poster: ContentVec: An Improved Self-Supervised Speech Representation by Disentangling Speakers »
Kaizhi Qian · Yang Zhang · Heting Gao · Junrui Ni · Cheng-I Lai · David Cox · Mark Hasegawa-Johnson · Shiyu Chang -
2022 Spotlight: Data-Efficient Double-Win Lottery Tickets from Robust Pre-training »
Tianlong Chen · Zhenyu Zhang · Sijia Liu · Yang Zhang · Shiyu Chang · Zhangyang “Atlas” Wang -
2022 Spotlight: ContentVec: An Improved Self-Supervised Speech Representation by Disentangling Speakers »
Kaizhi Qian · Yang Zhang · Heting Gao · Junrui Ni · Cheng-I Lai · David Cox · Mark Hasegawa-Johnson · Shiyu Chang -
2022 Spotlight: Antibody-Antigen Docking and Design via Hierarchical Structure Refinement »
Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2022 Spotlight: Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness »
Tianlong Chen · Huan Zhang · Zhenyu Zhang · Shiyu Chang · Sijia Liu · Pin-Yu Chen · Zhangyang “Atlas” Wang -
2022 Poster: Revisiting and Advancing Fast Adversarial Training Through The Lens of Bi-Level Optimization »
Yihua Zhang · Guanhua Zhang · Prashant Khanduri · Mingyi Hong · Shiyu Chang · Sijia Liu -
2022 Poster: Conformal Prediction Sets with Limited False Positives »
Adam Fisch · Tal Schuster · Tommi Jaakkola · Regina Barzilay -
2022 Poster: EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction »
Hannes Stärk · Octavian Ganea · Lagnajit Pattanaik · Regina Barzilay · Tommi Jaakkola -
2022 Spotlight: Revisiting and Advancing Fast Adversarial Training Through The Lens of Bi-Level Optimization »
Yihua Zhang · Guanhua Zhang · Prashant Khanduri · Mingyi Hong · Shiyu Chang · Sijia Liu -
2022 Spotlight: Conformal Prediction Sets with Limited False Positives »
Adam Fisch · Tal Schuster · Tommi Jaakkola · Regina Barzilay -
2022 Spotlight: EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction »
Hannes Stärk · Octavian Ganea · Lagnajit Pattanaik · Regina Barzilay · Tommi Jaakkola -
2022 Invited Talk: Solving the Right Problems: Making ML Models Relevant to Healthcare and the Life Sciences »
Regina Barzilay -
2021 Poster: Global Prosody Style Transfer Without Text Transcriptions »
Kaizhi Qian · Yang Zhang · Shiyu Chang · Jinjun Xiong · Chuang Gan · David Cox · Mark Hasegawa-Johnson -
2021 Oral: Global Prosody Style Transfer Without Text Transcriptions »
Kaizhi Qian · Yang Zhang · Shiyu Chang · Jinjun Xiong · Chuang Gan · David Cox · Mark Hasegawa-Johnson -
2021 Poster: Few-Shot Conformal Prediction with Auxiliary Tasks »
Adam Fisch · Tal Schuster · Tommi Jaakkola · Regina Barzilay -
2021 Poster: Predict then Interpolate: A Simple Algorithm to Learn Stable Classifiers »
Yujia Bao · Shiyu Chang · Regina Barzilay -
2021 Spotlight: Few-Shot Conformal Prediction with Auxiliary Tasks »
Adam Fisch · Tal Schuster · Tommi Jaakkola · Regina Barzilay -
2021 Spotlight: Predict then Interpolate: A Simple Algorithm to Learn Stable Classifiers »
Yujia Bao · Shiyu Chang · Regina Barzilay -
2020 Poster: Multi-Objective Molecule Generation using Interpretable Substructures »
Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2020 Poster: Educating Text Autoencoders: Latent Representation Guidance via Denoising »
Tianxiao Shen · Jonas Mueller · Regina Barzilay · Tommi Jaakkola -
2020 Poster: Invariant Rationalization »
Shiyu Chang · Yang Zhang · Mo Yu · Tommi Jaakkola -
2020 Poster: Proper Network Interpretability Helps Adversarial Robustness in Classification »
Akhilan Boopathy · Sijia Liu · Gaoyuan Zhang · Cynthia Liu · Pin-Yu Chen · Shiyu Chang · Luca Daniel -
2020 Poster: Hierarchical Generation of Molecular Graphs using Structural Motifs »
Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2020 Poster: Unsupervised Speech Decomposition via Triple Information Bottleneck »
Kaizhi Qian · Yang Zhang · Shiyu Chang · Mark Hasegawa-Johnson · David Cox -
2020 Poster: Improving Molecular Design by Stochastic Iterative Target Augmentation »
Kevin Yang · Wengong Jin · Kyle Swanson · Regina Barzilay · Tommi Jaakkola -
2019 Poster: AutoVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss »
Kaizhi Qian · Yang Zhang · Shiyu Chang · Xuesong Yang · Mark Hasegawa-Johnson -
2019 Oral: AutoVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss »
Kaizhi Qian · Yang Zhang · Shiyu Chang · Xuesong Yang · Mark Hasegawa-Johnson -
2018 Poster: Junction Tree Variational Autoencoder for Molecular Graph Generation »
Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2018 Oral: Junction Tree Variational Autoencoder for Molecular Graph Generation »
Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2017 Poster: Deriving Neural Architectures from Sequence and Graph Kernels »
Tao Lei · Wengong Jin · Regina Barzilay · Tommi Jaakkola -
2017 Talk: Deriving Neural Architectures from Sequence and Graph Kernels »
Tao Lei · Wengong Jin · Regina Barzilay · Tommi Jaakkola