Alex K. Park
Lead of AI at SOCAR
kp at socar dot krkyungho dot dev at gmail dot com
About
Hi, I'm the Product Owner and Lead of AI at SOCAR, the largest car-sharing platform in the Republic of Korea. I recently finished my MS in cybersecurity at Korea University, advised by Professor Huy Kang Kim. Before my MS, I was a visiting researcher at Carnegie Mellon University supported by the Ministry of Science and ICT of the Republic of Korea, and I completed my BA in Business Administration at Korea University Business School. My research interests are robustness in machine learning (especially in federated learning), representation learning, and utilization of causal inference to the interpretability of machine learning models.
News
July 2023 1 do a tech talk at Microsoft Startup Summit 2023
July 2023 1 paper accepted to EMNLP MRL Workshop 2023.
July 2023 1 paper accepted to ITSC 2023.
May 2023 My team hosts Foundational Model without Descriptive Caption (FMDC) Challenge in accordance with the VPLOW workshop in CVPR 2023.
Mar 2023 2 papers accepted to ICLR Workshop on Practical ML for Developing Countries (PML4DC) 2023.
Feb 2023 I do a tech talk at NAVER DEVIEW 2023, the largest S/W tech conference in Republic of Korea. Please visit the website for more detailed information.
Nov 2022 3 papers accepted to WACV Workshop on Dealing with Novelty in Open World (DNOW) 2023.
Oct 2022 1 paper accepted to NIPS Workshop on Meta-Learning (MetaLearn) 2022.
Sep 2022 1 paper accepted to ACCV 2022.
Jun 2022 1 paper accepted to DLP-KDD 2022, and another 1 paper is accepted to ITSC 2022.
April 2022 I do a keynote speech at 2022 AI Business Development Conference 2022. Please visit the website for more detailed information.
Feb 2022 1 paper is accepted to Computers & Security Journal, and 1 poster is accepted to NDSS 2022.
Jan 2022 Opened my Personal website!
Recent Works
Adapt and Prune Strategy for Multilingual Speech Foundational Model on Low-resourced Languages
Hyeonsoo Kim, Chunghyeon Cho, Hyejin Won, and Kyung Ho Park
EMNLP Multilingual Representation Learning (MRL) Workshop
[Paper] (To Be Updated)
Learning Spatiotemporal Representation in Nighttime Driving Video for Road Flood Detection
Euiin Yi, Hyunhee Chung, and Kyung Ho Park
ITSC 2023
[Paper] (To Be Updated)
PMixUp: Simultaneous Utilization of Part-of-Speech Replacement and Feature Space Interpolation for Text Data Augmentation
Hyeonsoo Kim, Hyejin Won, and Kyung Ho Park
ICLR Practical ML for Developing Countries Workshop 2023 (PML4DC 2023)
[Paper] (To Be Updated)
Uncovering the Effectiveness of Calibration on Open Intent Classification
Hyejin Won and Kyung Ho Park
ICLR Practical ML for Developing Countries Workshop 2023 (PML4DC 2023)
[Paper] (To Be Updated)
Phantom of Benchmark Dataset: Resolving Label Ambiguity Problem on Image Recognition in the Wild
Hyunhee Chung*, Kyung Ho Park*, Taewon Seo, and Seongwoo Cho
WACV Workshop on Dealing with Novelty in Open Worlds 2023
[Paper] (To Be Updated)
Is Meta-Learning Always Necessary?: A Practical ML Framework Solving Novel Tasks at Large-scale Car Sharing Platform
Hyunhee Chung* and Kyung Ho Park*
WACV Workshop on Dealing with Novelty in Open Worlds 2023
[Paper] (To Be Updated)
SOCAR: Socially-Obtained CAR Dataset for Image Recognition in the Wild
Taewon Seo*, Kyung Ho Park*, and Hyunhee Chung
WACV Workshop on Dealing with Novelty in Open Worlds 2023
[Paper] (To Be Updated)
Lightweight Prompt Learning with General Representation for Rehearsal-free Continual Learning
Hyunhee Chung and Kyung Ho Park
NIPS Workshop on Meta-Learning 2022
[Paper] (To Be Updated)
Shape Prior is Not All You Need: Discovering Balance between Texture and Shape bias in CNN
Hyunhee Chung* and Kyung Ho Park*
ACCV 2022
[Paper] (To Be Updated)
Cashflow Tracing: Detecting Online game bots leveraging financial analysis with Recurrent Neural Networks
Kyung Ho Park, Eunjo Lee, and Huy Kang Kim
CHI PLAY 2022 Work In Progress
[Paper] (To Be Updated)
Towards an Efficient ML System: Unveiling a Trade-off between Task Accuracy and Engineering Efficiency in a Large-scale Car Sharing Platform
Kyung Ho Park*, Hyunhee Chung* and Soonwoo Kwon
DLP-KDD (KDD Workshop) 2022
[Paper] (To Be Updated)
Did you Wash the Car?: Robust Car Wash Inspection System in a Large-scale Car Sharing Platform
Hyunhee Chung and Kyung Ho Park
ITSC 2022
[Paper] (To Be Updated)
Deep Dense Local Feature Matching and Vehicle Removal for Indoor Visual Localization
Kyung Ho Park
arXiv Preprint
[Paper]
Poster: A Good Representation Helps the Robustness of Federated Learning against Backdoor Attack
Kyung Ho Park*, Sanghoon Jeon*, and Huy Kang Kim
NDSS 2022
[Paper]
Publications and Patents
Traffic Accident Recognition in First-Person Videos by Learning a Spatio-Temporal Visual Pattern
Kyung Ho Park, Dong Hyun Ahn, and Huy Kang Kim
IEEE VTC 2021
[Paper]
Uncertainty Guided Pseudo-Labeling: Estimating Uncertainty on Ambiguous Data for Escalating Image Recognition Performance
Kyung Ho Park* and Hyunhee Chung*
ICAART 2021
[Paper] (To be updated)
Don't Wait Until the Accident Happens: Few-Shot Classification Framework for Car Accident Inspection in a Real World
Kyung Ho Park* and Hyunhee Chung*
ICIAP 2021
[Paper] (To be updated)
Visual Representation Learning for Automating Car Part Recognition in a Large-scale Car Sharing Platform
Kyung Ho Park, Yunhwan Kwon, Youngin Song, and Seongyun Byeon
IEEE CASE 2021
[Paper]
Stay As You Were!: Unsupervised Driver Behavior Profiling through Discovering Normality on Smartphone Sensor Measurements
Kyung Ho Park, Eunji Park, and Huy Kang Kim
IEEE ITSC 2021
[Paper]
Unsupervised Malicious Domain Detection with Less Labeling Effort
Kyung Ho Park, Hyun Min Song, Jeong Do Yoo, Kwangsoo Kim, Byungmo Cho, Suyoun Hong, and Huy Kang Kim
Computers and Security
[Paper] (To be updated)
Unsupervised Fault Detection on Unmanned Aerial Vehicles: Encoding and Thresholding Approach
Kyung Ho Park, Eunji Park, and Huy Kang Kim
Sensors
[Paper]
De-Wipimization: Detection of data wiping traces for investigating NTFS file system
Dong Bin Oh, Kyung Ho Park and Huy Kang Kim
Computers and Security
[Paper]
Unsupervised Driver Behavior Profiling leveraging Recurrent Neural Networks
Young Ah Choi*, Kyung Ho Park*, Eunji Park, and Huy Kang Kim
WISA 2021
[Paper]
Understand Watchdogs: Discover How Game Bot Get Discovered
Eunji Park, Kyung Ho Park and Huy Kang Kim
ICAART 2020
[Paper]
Unsupervised Intrusion Detection System for Unmanned Aerial Vehicle with Less Labeling Effort
Kyung Ho Park, Eunji Park and Huy Kang Kim
WISA 2020
[Paper]
This car is mine! Automobile theft countermeasure leveraging driver identification with generative adversarial networks
Kyung Ho Park and Huy Kang Kim
ESCAR Asia 2019
[Paper]
Automobile theft detection by clustering owner driver data
Yong Gu Kang, Kyung Ho Park and Huy Kang Kim
ESCAR Europe 2019
[Paper]
Show me your account: Detecting MMORPG Game Bots Leveraging Financial Analysis with LSTM
Kyung Ho Park, Eunjo Lee and Huy Kang Kim
WISA 2019
[Paper]
IoT network intrusion dataset
Hyunjae Kang, Dong Hyun Ahn, Jeong Do Yoo, Kyung Ho Park, and Huy Kang Kim
IEEE Dataport
[Dataset]
APPARATUS AND METHOD FOR DETECTING MALWARE COMMUNICATION WITH DGA GENERATED DOMAINS
Huy Kang Kim, Hyun Min Song, Jeong Do Yoo, Kyung Ho Park, Kwangsoo Kim, Byungmo Cho, and Suyoun Hong
Application Number 10-2020-0034699
Talks
March 2023 [Tech Talk] Towards New User Experience with LLMs
Talks Presented in Microsoft Startup Summit 2023
[Video]
March 2023 [Lecture] Towards Continuously Growing AI System with Continual Learning
Talks Presented in Korea Astronomy and Space Science Institute (KASI)
Februrary 2023 [Lecture] Towards Continuously Growing AI System with Continual Learning
Talks Presented in NAVER DEVIEW, 2023
[Website]
September 2022 [Lecture] Computer Vision and Natural Language Processing in SOCAR
Talks Presented in Google Machine Learning Bootcamp, 2022
[Website]
July 2022 [Lecture] From Research to Operational Excellence: How SOCAR Operates 20K Fleets with AI
Talks Presented in Deep Learning Summer School
[Website]
April 2022 [Keynote] From Research to Operational Excellence: How SOCAR Operates 20K Fleets with AI
Keynote Speech Presented in AI Business Development Conference 2022, COEX
[Website]
[Video] (Korean)
April 2022 [Lecture] AI Research 101: How to become an Applied Research Scientist?
Talks Presented in Korea University
[Slide] (Korean, To be released)
Feb 2022 [Lecture] AI Research in SOCAR
Talks Presented in Pukyong National University
[Slide] (Korean)
Oct 2021 [Tech Talk] Towards Efficient Real World Image Recognition Model leveraging Transfer Learning
Talks Presented in Datayanolja, Data Conference in Republic of Korea
[Video] (Korean)
Media Release
October 2023 쏘카 "실수 안하는 'AI 상담원' 만들겠다
URL (Korean)
October 2023 스타트업이 챗GPT로 사업하는 노하우 대방출
URL (Korean)
May 2023 'AI 퍼스트' 쏘카, 연내 AI 고객센터 솔루션 도입
URL (Korean)
February 2023 [데뷰23] AI모델 효과적으로 개선하려면
URL (Korean)
August 2022 쏘카, 글로벌 학회서 '차량 오염·파손 탐지' AI 기술 소개
URL (Korean)
July 2022 [AI활용사례] 우리 비즈니스에 굳이 AI가 필요할까
URL (Korean)
June 2022 [직무 인터뷰] AI 기술로 더 가치 있는 세상, 혁신적인 미래를 열다, 쏘카 데이터비즈니스본부 AI팀 박경호 리더
URL (Korean)
April 2022 [쏘카 테크블로그] 딥러닝 기반 세차 인증 자동화 모델 개발 이야기
URL (Korean)
April 2022 [쏘카 테크블로그] 쏘카 AI팀의 Applied Research Scientist는 어떤 일을 하나요?
URL (Korean)
May 2021 쏘카, '카메라+AI' 기술로 車 사고 예측·알림·추적한다
URL (Korean)
Oct 2021 [카&테크]자동차 파손·결함 잡아내는 비전 AI 기술
URL (Korean)
Oct 2021 코로나19 이후 밀집 지역 벗어나 탈수도권 이동 가속화
URL (Korean)
Lectures
[Korea Statistics (통계청)] Datascience Intensive Course
August, 2023