Author name cluster

Jaeho Lee

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

11 papers

2 author rows

TMLR Journal 2023 Journal Article

Breaking the Spurious Causality of Conditional Generation via Fairness Intervention with Corrective Sampling

Junhyun Nam
Sangwoo Mo
Jaeho Lee
Jinwoo Shin

Trying to capture the sample-label relationship, conditional generative models often end up inheriting the spurious correlation in the training dataset, giving label-conditional distributions that are severely imbalanced in another latent attribute. To mitigate such undesirable correlations engraved into generative models, which we call spurious causality, we propose a general two-step strategy. (a) Fairness Intervention (FI): Emphasize the minority samples that are hard to be generated due to the spurious correlation in the training dataset. (b) Corrective Sampling (CS): Filter the generated samples explicitly to follow the desired label-conditional latent attribute distribution. We design the fairness intervention for various degrees of supervision on the spurious attribute, including unsupervised, weakly-supervised, and semi-supervised scenarios. Our experimental results show that the proposed FICS can successfully resolve the spurious correlation in generated samples on various datasets.

PDF Details

NeurIPS Conference 2023 Conference Paper

Learning Large-scale Neural Fields via Context Pruned Meta-Learning

Jihoon Tack
Subin Kim
Sihyun Yu
Jaeho Lee
Jinwoo Shin
Jonathan Richard Schwarz

We introduce an efficient optimization-based meta-learning technique for large-scale neural field training by realizing significant memory savings through automated online context point selection. This is achieved by focusing each learning step on the subset of data with the highest expected immediate improvement in model quality, resulting in the almost instantaneous modeling of global structure and subsequent refinement of high-frequency details. We further improve the quality of our meta-learned initialization by introducing a bootstrap correction resulting in the minimization of any error introduced by reduced context sets while simultaneously mitigating the well-known myopia of optimization-based meta-learning. Finally, we show how gradient re-scaling at meta-test time allows the learning of extremely high-quality neural fields in significantly shortened optimization procedures. Our framework is model-agnostic, intuitive, straightforward to implement, and shows significant reconstruction improvements for a wide range of signals. We provide an extensive empirical evaluation on nine datasets across multiple multiple modalities, demonstrating state-of-the-art results while providing additional insight through careful analysis of the algorithmic components constituting our method. Code is available at https: //github. com/jihoontack/GradNCP

PDF Details

NeurIPS Conference 2022 Conference Paper

Meta-Learning with Self-Improving Momentum Target

Jihoon Tack
Jongjin Park
Hankook Lee
Jaeho Lee
Jinwoo Shin

The idea of using a separately trained target model (or teacher) to improve the performance of the student model has been increasingly popular in various machine learning domains, and meta-learning is no exception; a recent discovery shows that utilizing task-wise target models can significantly boost the generalization performance. However, obtaining a target model for each task can be highly expensive, especially when the number of tasks for meta-learning is large. To tackle this issue, we propose a simple yet effective method, coined Self-improving Momentum Target (SiMT). SiMT generates the target model by adapting from the temporal ensemble of the meta-learner, i. e. , the momentum network. This momentum network and its task-specific adaptations enjoy a favorable generalization performance, enabling self-improving of the meta-learner through knowledge distillation. Moreover, we found that perturbing parameters of the meta-learner, e. g. , dropout, further stabilize this self-improving process by preventing fast convergence of the distillation loss during meta-training. Our experimental results demonstrate that SiMT brings a significant performance gain when combined with a wide range of meta-learning methods under various applications, including few-shot regression, few-shot classification, and meta-reinforcement learning. Code is available at https: //github. com/jihoontack/SiMT.

PDF Details

NeurIPS Conference 2022 Conference Paper

Scalable Neural Video Representations with Learnable Positional Features

Subin Kim
Sihyun Yu
Jaeho Lee
Jinwoo Shin

Succinct representation of complex signals using coordinate-based neural representations (CNRs) has seen great progress, and several recent efforts focus on extending them for handling videos. Here, the main challenge is how to (a) alleviate a compute-inefficiency in training CNRs to (b) achieve high-quality video encoding while (c) maintaining the parameter-efficiency. To meet all requirements (a), (b), and (c) simultaneously, we propose neural video representations with learnable positional features (NVP), a novel CNR by introducing "learnable positional features" that effectively amortize a video as latent codes. Specifically, we first present a CNR architecture based on designing 2D latent keyframes to learn the common video contents across each spatio-temporal axis, which dramatically improves all of those three requirements. Then, we propose to utilize existing powerful image and video codecs as a compute-/memory-efficient compression procedure of latent codes. We demonstrate the superiority of NVP on the popular UVG benchmark; compared with prior arts, NVP not only trains 2 times faster (less than 5 minutes) but also exceeds their encoding quality as 34. 07$\rightarrow$34. 57 (measured with the PSNR metric), even using $>$8 times fewer parameters. We also show intriguing properties of NVP, e. g. , video inpainting, video frame interpolation, etc.

PDF Details

ICRA Conference 2021 Conference Paper

An integrated approach for determining objects to be relocated and their goal positions inside clutter for object retrieval

Jeeho Ahn
Jaeho Lee
SangHun Cheong
ChangHwan Kim 0002
Changjoo Nam

We consider the problem of rearranging objects in a cluttered and confined space using a robotic manipulator. The goal is to retrieve a target object from the clutter where the target is occluded by other objects. In situations where overhand grasps are not allowed, the robot needs to remove some objects to make the target accessible. In the course of removing the objects, the robot also needs to determine the locations to place the removed objects. If the robot can access enough empty spaces around or inside the clutter, the placement of the objects is trivially simple. If empty spaces are scarce, placing objects should be done in a principled way as an incorrect placement would deplete the empty spaces quickly. In this work, we propose a method that solves the problems of what and where to relocate objects inside the clutter to retrieve the target. Previously, there have been several efficient methods proposed that deal with each of the what and where to relocate problems separately. We solve the problems together using a graph structure constructed from an object configuration. Also, the method runs fast so scalable in the number of objects. Compared to a state-of-the-art method, our method reduces task and motion planning time up to 74. 9% (at least 56. 7%) and has a higher success rate under a short time limit for planning, which is 3 minutes.

Details

AAAI Conference 2021 Conference Paper

MASKER: Masked Keyword Regularization for Reliable Text Classification

Seung Jun Moon
Sangwoo Mo
Kimin Lee
Jaeho Lee
Jinwoo Shin

Pre-trained language models have achieved state-of-the-art accuracies on various text classification tasks, e. g. , sentiment analysis, natural language inference, and semantic textual similarity. However, the reliability of the fine-tuned text classifiers is an often underlooked performance criterion. For instance, one may desire a model that can detect out-of-distribution (OOD) samples (drawn far from training distribution) or be robust against domain shifts. We claim that one central obstacle to the reliability is the over-reliance of the model on a limited number of keywords, instead of looking at the whole context. In particular, we find that (a) OOD samples often contain indistribution keywords, while (b) cross-domain samples may not always contain keywords; over-relying on the keywords can be problematic for both cases. In light of this observation, we propose a simple yet effective fine-tuning method, coined masked keyword regularization (MASKER), that facilitates context-based prediction. MASKER regularizes the model to reconstruct the keywords from the rest of the words and make low-confidence predictions without enough context. When applied to various pre-trained language models (e. g. , BERT, RoBERTa, and ALBERT), we demonstrate that MASKER improves OOD detection and cross-domain generalization without degrading classification accuracy. Code is available at https: //github. com/alinlab/MASKER.

PDF Details

NeurIPS Conference 2021 Conference Paper

Meta-Learning Sparse Implicit Neural Representations

Jaeho Lee
Jihoon Tack
Namhoon Lee
Jinwoo Shin

Implicit neural representations are a promising new avenue of representing general signals by learning a continuous function that, parameterized as a neural network, maps the domain of a signal to its codomain; the mapping from spatial coordinates of an image to its pixel values, for example. Being capable of conveying fine details in a high dimensional signal, unboundedly of its domain, implicit neural representations ensure many advantages over conventional discrete representations. However, the current approach is difficult to scale for a large number of signals or a data set, since learning a neural representation---which is parameter heavy by itself---for each signal individually requires a lot of memory and computations. To address this issue, we propose to leverage a meta-learning approach in combination with network compression under a sparsity constraint, such that it renders a well-initialized sparse parameterization that evolves quickly to represent a set of unseen signals in the subsequent training. We empirically demonstrate that meta-learned sparse neural representations achieve a much smaller loss than dense meta-learned models with the same number of parameters, when trained to fit each signal using the same number of optimization steps.

PDF Details

NeurIPS Conference 2020 Conference Paper

Learning Bounds for Risk-sensitive Learning

Jaeho Lee
Sejun Park
Jinwoo Shin

In risk-sensitive learning, one aims to find a hypothesis that minimizes a risk-averse (or risk-seeking) measure of loss, instead of the standard expected loss. In this paper, we propose to study the generalization properties of risk-sensitive learning schemes whose optimand is described via optimized certainty equivalents (OCE): our general scheme can handle various known risks, e. g. , the entropic risk, mean-variance, and conditional value-at-risk, as special cases. We provide two learning bounds on the performance of empirical OCE minimizer. The first result gives an OCE guarantee based on the Rademacher average of the hypothesis space, which generalizes and improves existing results on the expected loss and the conditional value-at-risk. The second result, based on a novel variance-based characterization of OCE, gives an expected loss guarantee with a suppressed dependence on the smoothness of the selected OCE. Finally, we demonstrate the practical implications of the proposed bounds via exploratory experiments on neural networks.

PDF Details

NeurIPS Conference 2020 Conference Paper

Learning from Failure: De-biasing Classifier from Biased Classifier

Junhyun Nam
Hyuntak Cha
Sungsoo Ahn
Jaeho Lee
Jinwoo Shin

Neural networks often learn to make predictions that overly rely on spurious corre- lation existing in the dataset, which causes the model to be biased. While previous work tackles this issue by using explicit labeling on the spuriously correlated attributes or presuming a particular bias type, we instead utilize a cheaper, yet generic form of human knowledge, which can be widely applicable to various types of bias. We first observe that neural networks learn to rely on the spurious correlation only when it is “easier” to learn than the desired knowledge, and such reliance is most prominent during the early phase of training. Based on the obser- vations, we propose a failure-based debiasing scheme by training a pair of neural networks simultaneously. Our main idea is twofold; (a) we intentionally train the first network to be biased by repeatedly amplifying its “prejudice”, and (b) we debias the training of the second network by focusing on samples that go against the prejudice of the biased network in (a). Extensive experiments demonstrate that our method significantly improves the training of network against various types of biases in both synthetic and real-world datasets. Surprisingly, our framework even occasionally outperforms the debiasing methods requiring explicit supervision of the spuriously correlated attributes.

PDF Details

NeurIPS Conference 2018 Conference Paper

Minimax Statistical Learning with Wasserstein distances

Jaeho Lee
Maxim Raginsky

As opposed to standard empirical risk minimization (ERM), distributionally robust optimization aims to minimize the worst-case risk over a larger ambiguity set containing the original empirical distribution of the training data. In this work, we describe a minimax framework for statistical learning with ambiguity sets given by balls in Wasserstein space. In particular, we prove generalization bounds that involve the covering number properties of the original ERM problem. As an illustrative example, we provide generalization guarantees for transport-based domain adaptation problems where the Wasserstein distance between the source and target domain distributions can be reliably estimated from unlabeled samples.

PDF Details

AAAI Conference 1994 Conference Paper

Structured Circuit Semantics for Reactive Plan Execution Systems

Jaeho Lee

A variety of reactive plan execution systems have been developed in recent years, each attempting to solve the problem of taking reasonable courses of action fast enough in a dynamically changing world. Comparing these competing approaches, and collecting the best features of each, has been problematic because of the diverse representations and (sometimes implicit) control structures that they have employed. To rectify this problem, we have extended the circuit semantics notion of teleo-reactive programs into richer, yet compact semantics, called structured circuit semantics (SCS), that can be used to explicitly represent the control behavior of various reactive execution systems. By transforming existing systems into SCS, we can identify underlying control assumptions and begin to identify more rigorously the strengths and limitations of these systems. Moreover, SCS provides a basis for constructing new reactive execution systems, with more understandable semantics, that can be tailored to particular domain needs.

PDF Details