Timezone: »
Most existing set encoding algorithms operate under the implicit assumption that all the set elements are accessible, and that there are ample computational and memory resources to load the set into memory during training and inference. However, both assumptions fail when the set is excessively large such that it is impossible to load all set elements into memory, or when data arrives in a stream. To tackle such practical challenges in large-scale set encoding, the general set-function constraints of permutation invariance and equivariance are not sufficient. We introduce a new property termed Mini-Batch Consistency (MBC) that is required for large scale mini-batch set encoding. Additionally, we present a scalable and efficient attention-based set encoding mechanism that is amenable to mini-batch processing of sets, and capable of updating set representations as data arrives. The proposed method adheres to the required symmetries of invariance and equivariance as well as maintaining MBC for any partition of the input set. We perform extensive experiments and show that our method is computationally efficient and results in rich set encoding representations for set-structured data.
Author Information
Bruno Andreis (KAIST)
Jeffrey Willette (Korea Advanced Institute of Science and Technology)
Juho Lee (KAIST, AITRICS)
Sung Ju Hwang (KAIST, AITRICS)
More from the Same Authors
-
2021 Spotlight: Hardware-adaptive Efficient Latency Prediction for NAS via Meta-Learning »
Hayeon Lee · Sewoong Lee · Song Chong · Sung Ju Hwang -
2021 Spotlight: Task-Adaptive Neural Network Search with Meta-Contrastive Learning »
Wonyong Jeong · Hayeon Lee · Geon Park · Eunyoung Hyung · Jinheon Baek · Sung Ju Hwang -
2021 : Skill-based Meta-Reinforcement Learning »
Taewook Nam · Shao-Hua Sun · Karl Pertsch · Sung Ju Hwang · Joseph Lim -
2021 : Skill-based Meta-Reinforcement Learning »
Taewook Nam · Shao-Hua Sun · Karl Pertsch · Sung Ju Hwang · Joseph Lim -
2022 Poster: Learning to Generate Inversion-Resistant Model Explanations »
Hoyong Jeong · Suyoung Lee · Sung Ju Hwang · Sooel Son -
2022 : Distortion-Aware Network Pruning and Feature Reuse for Real-time Video Segmentation »
Hyunsu Rhee · Dongchan Min · Sunil Hwang · Bruno Andreis · Sung Ju Hwang -
2022 : SPRINT: Scalable Semantic Policy Pre-training via Language Instruction Relabeling »
Jesse Zhang · Karl Pertsch · Jiahui Zhang · Taewook Nam · Sung Ju Hwang · Xiang Ren · Joseph Lim -
2022 : SPRINT: Scalable Semantic Policy Pre-training via Language Instruction Relabeling »
Jesse Zhang · Karl Pertsch · Jiahui Zhang · Taewook Nam · Sung Ju Hwang · Xiang Ren · Joseph Lim -
2023 Poster: Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks »
Minki Kang · Seanie Lee · Jinheon Baek · Kenji Kawaguchi · Sung Ju Hwang -
2023 Poster: Generalizable Lightweight Proxy for Robust NAS against Diverse Perturbations »
Hyeonjeong Ha · Minseon Kim · Sung Ju Hwang -
2023 Poster: STXD: Structural and Temporal Cross-Modal Distillation for Multi-View 3D Object Detection »
Sujin Jang · Dae Ung Jo · Sung Ju Hwang · Dongwook Lee · Daehyun Ji -
2023 Poster: Effective Targeted Attacks for Adversarial Self-Supervised Learning »
Minseon Kim · Hyeonjeong Ha · Sooel Son · Sung Ju Hwang -
2022 Poster: Factorized-FL: Personalized Federated Learning with Parameter Factorization & Similarity Matching »
Wonyong Jeong · Sung Ju Hwang -
2022 Poster: Graph Self-supervised Learning with Accurate Discrepancy Learning »
Dongki Kim · Jinheon Baek · Sung Ju Hwang -
2022 Poster: Set-based Meta-Interpolation for Few-Task Meta-Learning »
Seanie Lee · Bruno Andreis · Kenji Kawaguchi · Juho Lee · Sung Ju Hwang -
2021 Poster: Edge Representation Learning with Hypergraphs »
Jaehyeong Jo · Jinheon Baek · Seul Lee · Dongki Kim · Minki Kang · Sung Ju Hwang -
2021 Poster: Hit and Lead Discovery with Explorative RL and Fragment-based Molecule Generation »
Soojung Yang · Doyeong Hwang · Seul Lee · Seongok Ryu · Sung Ju Hwang -
2021 Poster: Diversity Matters When Learning From Ensembles »
Giung Nam · Jongmin Yoon · Yoonho Lee · Juho Lee -
2021 Poster: Hardware-adaptive Efficient Latency Prediction for NAS via Meta-Learning »
Hayeon Lee · Sewoong Lee · Song Chong · Sung Ju Hwang -
2021 Poster: Task-Adaptive Neural Network Search with Meta-Contrastive Learning »
Wonyong Jeong · Hayeon Lee · Geon Park · Eunyoung Hyung · Jinheon Baek · Sung Ju Hwang -
2020 Poster: Bootstrapping neural processes »
Juho Lee · Yoonho Lee · Jungtaek Kim · Eunho Yang · Sung Ju Hwang · Yee Whye Teh -
2020 Poster: Neural Complexity Measures »
Yoonho Lee · Juho Lee · Sung Ju Hwang · Eunho Yang · Seungjin Choi