arXiv subjects

Sundong Kim

Publications and source records attributed to Sundong Kim.

34 records · Page 2Linked to original sources

arXiv2023

Customs Import Declaration Datasets

Given the huge volume of cross-border flows, effective and efficient control of trade becomes more crucial in protecting people and society from illicit trade. However, limited accessibility of the transaction-level trade datasets hinders the progress of open research, and lots of customs administrations have not benefited from the recent progress in data-based risk management. In this paper, we introduce an import declaration dataset to facilitate the collaboration between domain experts in customs administrations and researchers from diverse domains, such as data science and machine learning. The dataset contains 54,000 artificially generated trades with 22 key attributes, and it is synthesized with conditional tabular GAN while maintaining correlated features. Synthetic data has several advantages. First, releasing the dataset is free from restrictions that do not allow disclosing the original import data. The fabrication step minimizes the possible identity risk which may exist in trade statistics. Second, the published data follow a similar distribution to the source data so that it can be used in various downstream tasks. Hence, our dataset can be used as a benchmark for testing the performance of any classification algorithm. With the provision of data and its generation process, we open baseline codes for fraud detection tasks, as we empirically show that more advanced algorithms can better detect fraud.

cs.LG↗

arXiv2023

Towards Attack-tolerant Federated Learning via Critical Parameter Analysis

Federated learning is used to train a shared model in a decentralized way without clients sharing private data with each other. Federated learning systems are susceptible to poisoning attacks when malicious clients send false updates to the central server. Existing defense strategies are ineffective under non-IID data settings. This paper proposes a new defense strategy, FedCPA (Federated learning with Critical Parameter Analysis). Our attack-tolerant aggregation method is based on the observation that benign local models have similar sets of top-k and bottom-k critical parameters, whereas poisoned local models do not. Experiments with different attack scenarios on multiple datasets demonstrate that our model outperforms existing defense strategies in defending against poisoning attacks.

cs.LG↗

arXiv2023

FedDefender: Client-Side Attack-Tolerant Federated Learning

Federated learning enables learning from decentralized data sources without compromising privacy, which makes it a crucial technique. However, it is vulnerable to model poisoning attacks, where malicious clients interfere with the training process. Previous defense mechanisms have focused on the server-side by using careful model aggregation, but this may not be effective when the data is not identically distributed or when attackers can access the information of benign clients. In this paper, we propose a new defense mechanism that focuses on the client-side, called FedDefender, to help benign clients train robust local models and avoid the adverse impact of malicious model updates from attackers, even when a server-side defense cannot identify or remove adversaries. Our method consists of two main components: (1) attack-tolerant local meta update and (2) attack-tolerant global knowledge distillation. These components are used to find noise-resilient model parameters while accurately extracting knowledge from a potentially corrupted global model. Our client-side defense strategy has a flexible structure and can work in conjunction with any existing server-side strategies. Evaluations of real-world scenarios across multiple datasets show that the proposed method enhances the robustness of federated learning against model poisoning attacks.

cs.CR↗

arXiv2023

Unraveling the ARC Puzzle: Mimicking Human Solutions with Object-Centric Decision Transformer

In the pursuit of artificial general intelligence (AGI), we tackle Abstraction and Reasoning Corpus (ARC) tasks using a novel two-pronged approach. We employ the Decision Transformer in an imitation learning paradigm to model human problem-solving, and introduce an object detection algorithm, the Push and Pull clustering method. This dual strategy enhances AI's ARC problem-solving skills and provides insights for AGI progression. Yet, our work reveals the need for advanced data collection tools, robust training datasets, and refined model structures. This study highlights potential improvements for Decision Transformers and propels future AGI research.

cs.AI↗

arXiv2023

DualFair: Fair Representation Learning at Both Group and Individual Levels via Contrastive Self-supervision

Algorithmic fairness has become an important machine learning problem, especially for mission-critical Web applications. This work presents a self-supervised model, called DualFair, that can debias sensitive attributes like gender and race from learned representations. Unlike existing models that target a single type of fairness, our model jointly optimizes for two fairness criteria - group fairness and counterfactual fairness - and hence makes fairer predictions at both the group and individual levels. Our model uses contrastive loss to generate embeddings that are indistinguishable for each protected group, while forcing the embeddings of counterfactual pairs to be similar. It then uses a self-knowledge distillation method to maintain the quality of representation for the downstream tasks. Extensive analysis over multiple datasets confirms the model's validity and further shows the synergy of jointly addressing two fairness criteria, suggesting the model's potential value in fair intelligent Web applications.

cs.LG↗

arXiv2022

FedX: Unsupervised Federated Learning with Cross Knowledge Distillation

This paper presents FedX, an unsupervised federated learning framework. Our model learns unbiased representation from decentralized and heterogeneous local data. It employs a two-sided knowledge distillation with contrastive learning as a core component, allowing the federated system to function without requiring clients to share any data features. Furthermore, its adaptable architecture can be used as an add-on module for existing unsupervised algorithms in federated settings. Experiments show that our model improves performance significantly (1.58--5.52pp) on five unsupervised algorithms.

cs.CV↗

arXiv2022

Active Learning for Human-in-the-Loop Customs Inspection

We study the human-in-the-loop customs inspection scenario, where an AI-assisted algorithm supports customs officers by recommending a set of imported goods to be inspected. If the inspected items are fraudulent, the officers can levy extra duties. Th formed logs are then used as additional training data for successive iterations. Choosing to inspect suspicious items first leads to an immediate gain in customs revenue, yet such inspections may not bring new insights for learning dynamic traffic patterns. On the other hand, inspecting uncertain items can help acquire new knowledge, which will be used as a supplementary training resource to update the selection systems. Based on multiyear customs datasets obtained from three countries, we demonstrate that some degree of exploration is necessary to cope with domain shifts in trade data. The results show that a hybrid strategy of selecting likely fraudulent and uncertain items will eventually outperform the exploitation-only strategy.

cs.LG↗

arXiv2022

Knowledge Sharing via Domain Adaptation in Customs Fraud Detection

Knowledge of the changing traffic is critical in risk management. Customs offices worldwide have traditionally relied on local resources to accumulate knowledge and detect tax fraud. This naturally poses countries with weak infrastructure to become tax havens of potentially illicit trades. The current paper proposes DAS, a memory bank platform to facilitate knowledge sharing across multi-national customs administrations to support each other. We propose a domain adaptation method to share transferable knowledge of frauds as prototypes while safeguarding the local trade information. Data encompassing over 8 million import declarations have been used to test the feasibility of this new system, which shows that participating countries may benefit up to 2-11 times in fraud detection with the help of shared knowledge. We discuss implications for substantial tax revenue potential and strengthened policy against illicit trades.

cs.AI↗

arXiv2021

Classification of Goods Using Text Descriptions With Sentences Retrieval

The task of assigning and validating internationally accepted commodity code (HS code) to traded goods is one of the critical functions at the customs office. This decision is crucial to importers and exporters, as it determines the tariff rate. However, similar to court decisions made by judges, the task can be non-trivial even for experienced customs officers. The current paper proposes a deep learning model to assist this seemingly challenging HS code classification. Together with Korea Customs Service, we built a decision model based on KoELECTRA that suggests the most likely heading and subheadings (i.e., the first four and six digits) of the HS code. Evaluation on 129,084 past cases shows that the top-3 suggestions made by our model have an accuracy of 95.5% in classifying 265 subheadings. This promising result implies algorithms may reduce the time and effort taken by customs officers substantially by assisting the HS code classification task.

cs.AI↗

arXiv2021

Customs Fraud Detection in the Presence of Concept Drift

Capturing the changing trade pattern is critical in customs fraud detection. As new goods are imported and novel frauds arise, a drift-aware fraud detection system is needed to detect both known frauds and unknown frauds within a limited budget. The current paper proposes ADAPT, an adaptive selection method that controls the balance between exploitation and exploration strategies used for customs fraud detection. ADAPT makes use of the model performance trends and the amount of concept drift to determine the best exploration ratio at every time. Experiments on data from four countries over several years show that each country requires a different amount of exploration for maintaining its fraud detection system. We find the system with ADAPT can gradually adapt to the dataset and find the appropriate amount of exploration ratio with high performance.

cs.AI↗

arXiv2021

Responses to COVID-19 with Probabilistic Programming

The COVID-19 pandemic left its unique mark on the 21st century as one of the most significant disasters in history, triggering governments all over the world to respond with a wide range of interventions. However, these restrictions come with a substantial price tag. It is crucial for governments to form anti-virus strategies that balance the trade-off between protecting public health and minimizing the economic cost. This work proposes a probabilistic programming method to quantify the efficiency of major non-pharmaceutical interventions. We present a generative simulation model that accounts for the economic and human capital cost of adopting such strategies, and provide an end-to-end pipeline to simulate the virus spread and the incurred loss of various policy combinations. By investigating the national response in 10 countries covering four continents, we found that social distancing coupled with contact tracing is the most successful policy, reducing the virus transmission rate by 96\% along with a 98\% reduction in economic and human capital loss. Together with experimental results, we open-sourced a framework to test the efficacy of each policy combination.

cs.CE↗

arXiv2021

Improving Unsupervised Image Clustering With Robust Learning

Unsupervised image clustering methods often introduce alternative objectives to indirectly train the model and are subject to faulty predictions and overconfident results. To overcome these challenges, the current research proposes an innovative model RUC that is inspired by robust learning. RUC's novelty is at utilizing pseudo-labels of existing image clustering models as a noisy dataset that may include misclassified samples. Its retraining process can revise misaligned knowledge and alleviate the overconfidence problem in predictions. The model's flexible structure makes it possible to be used as an add-on module to other clustering methods and helps them achieve better performance on multiple datasets. Extensive experiments show that the proposed model can adjust the model confidence with better calibration and gain additional robustness against adversarial noise.

cs.CV↗

arXiv2020

Carpe Diem, Seize the Samples Uncertain "At the Moment" for Adaptive Batch Selection

The accuracy of deep neural networks is significantly affected by how well mini-batches are constructed during the training step. In this paper, we propose a novel adaptive batch selection algorithm called Recency Bias that exploits the uncertain samples predicted inconsistently in recent iterations. The historical label predictions of each training sample are used to evaluate its predictive uncertainty within a sliding window. Then, the sampling probability for the next mini-batch is assigned to each training sample in proportion to its predictive uncertainty. By taking advantage of this design, Recency Bias not only accelerates the training step but also achieves a more accurate network. We demonstrate the superiority of Recency Bias by extensive evaluation on two independent tasks. Compared with existing batch selection methods, the results showed that Recency Bias reduced the test error by up to 20.97% in a fixed wall-clock training time. At the same time, it improved the training time by up to 59.32% to reach the same test error

cs.LG↗

arXiv2020

Disruption in the Chinese E-Commerce During COVID-19

The recent outbreak of the novel coronavirus (COVID-19) has infected millions of citizens worldwide and claimed many lives. This paper examines its impact on the Chinese e-commerce market by analyzing behavioral changes seen from a large online shopping platform. We first conduct a time series analysis to identify product categories that faced the most extensive disruptions. The time-lagged analysis shows that behavioral patterns seen in shopping actions are highly responsive to epidemic development. Based on these findings, we present a consumer demand prediction method by encompassing the epidemic statistics and behavioral features for COVID-19 related products. Experiment results demonstrate that our predictions outperform existing baselines and further extend to the long-term and province-level forecasts. We discuss how our market analysis and prediction can help better prepare for future pandemics by gaining an extra time to launch preventive steps.

physics.soc-ph↗

arXiv2016

Behavior of Self-Motivated Agents in Complex Networks

Traditional evolutionary game theory describes how certain strategy spreads throughout the system where individual player imitates the most successful strategy among its neighborhood. Accordingly, player doesn't have own authority to change their state. However in the human society, peoples do not just follow strategies of other people, they choose their own strategy. In order to see the decision of each agent in timely basis and differentiate between network structures, we conducted multi-agent based modeling and simulation. In this paper, agent can decide its own strategy by payoff comparison and we name this agent as "Self-motivated agent". To explain the behavior of self-motivated agent, prisoner's dilemma game with cooperator, defector, loner and punisher are considered as an illustrative example. We performed simulation by differentiating participation rate, mutation rate and the degree of network, and found the special coexisting conditions.

cs.MA↗

arXiv2016

Automatic Knowledge Base Evolution by Learning Instances

Knowledge base is the way to store structured and unstructured data throughout the web. Since the size of the web is increasing rapidly, there are huge needs to structure the knowledge in a fully automated way. However fully-automated knowledge-base evolution on the Semantic Web is a major challenges, although there are many ontology evolution techniques available. Therefore learning ontology automatically can contribute to the semantic web society significantly. In this paper, we propose full-automated ontology learning algorithm to generate refined knowledge base from incomplete knowledge base and rdf-triples. Our algorithm is data-driven approach which is based on the property of each instance. Ontology class is being elaborated by generalizing frequent property of its instances. By using that developed class information, each instance can find its most relatively matching class. By repeating these two steps, we achieve fully-automated ontology evolution from incomplete basic knowledge base.

cs.AI↗