2022 South Korea Finalists

Application number	Paper title	Student(s)	University
QIFK2022_1	TOWARDS CONTINUAL KNOWLEDGE LEARNING OF LANGUAGE MODELS	Joel Jang	KAIST
QIFK2022_4	Guaranteeing Safety Despite Physical Errors in Cyber-Physical Systems	Jongwoo-Han	Seoul National University
QIFK2022_9	UDA-COPE: Unsupervised Domain Adaptation for Category-level Object Pose Estimation	Taeyeop Lee	KAIST
QIFK2022_10	DASO: Distribution-Aware Semantics-Oriented Pseudo-label for Imbalanced Semi-Supervised Learning	Youngtaek Oh	KAIST
QIFK2022_11	Reflection and Rotation Symmetry Detection via Equivariant Learning	Ahyun Seo	POSTECH
QIFK2022_25	UBoCo : Unsupervised Boundary Contrastive Learning for Generic Event Boundary Detection	Hyolim Kang, Jinwoo Kim	Yonsei University
QIFK2022_27	Camera Distortion-aware 3D Human Pose Estimation in Video with Optimization-based Meta-Learning	Hanbyel Cho	KAIST
QIFK2022_37	Query-Efficient and Scalable Black-Box Adversarial Attacks on Discrete Sequential Data via Bayesian Optimization	Deokjae-Lee	Seoul National University
QIFK2022_42	Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation	Donghyeon Baek, Youngmin Oh	Yonsei University
QIFK2022_52	Learn-Explain-Reinforce: Counterfactual Reasoning and Its Guidance to Reinforce an Alzheimer’s Disease Diagnosis Model	Kwanseok Oh, Jee Seok Yoon	Korea University
QIFK2022_58	MASER: Multi-Agent Reinforcement Learning with Subgoals Generated from Experience Replay Buffer	Jeewon-Jeon	KAIST
QIFK2022_66	A Time-to-first-spike Coding and Conversion Aware Training for Energy-Efficient Deep Spiking Neural Network Processor Design	Dongwoo Lew	Korea University
QIFK2022_68	Robust Continual Test-time Adaptation: Instance-aware BN and Prediction-balanced Memory	Taesik Gong	KAIST
QIFK2022_69	Correlation Verification for Image Retrieval	Seongwon Lee	Yonsei University
QIFK2022_71	Uncertainty-Based Offline Reinforcement Learning with Diversified Q-Ensemble	Gaon An, Seungyong Moon	Seoul National University
QIFK2022_72	Non-Probability Sampling Network for Stochastic Human Trajectory Prediction	Inhwan-Bae	GIST
QIFK2022_74	PointMixer: MLP-Mixer for Point Cloud Understanding	Jaesung Choe, Chunghyun Park	KAIST, POSTECH
QIFK2022_75	Dataset Condensation with Contrastive Signals	Saehyung-Lee	Seoul National University
QIFK2022_76	Self-Taught Metric Learning without Labels	Sungyeon Kim	POSTECH
QIFK2022_80	DSPU: A 281.6mW Real-Time Depth Signal Processing Unit for Deep Learning-Based Dense RGB-D Data Acquisition with Depth Fusion and 3D Bounding Box Extraction in Mobile Platforms	Dongseok Im	KAIST
QIFK2022_81	Preemptive Image Robustification for Protecting Users against Man-in-the-Middle Adversarial Attacks	Seungyong Moon, Gaon An	Seoul National University
QIFK2022_83	WHAT MAKES BETTER AUGMENTATION STRATEGIES? AUGMENT DIFFICULT BUT NOT TOO DIFFERENT	Jaehyung Kim	KAIST
QIFK2022_85	One-Trimap Video Matting	Hongje Seong	Yonsei University
QIFK2022_86	Fast Point Transformer	Chunghyun Park	POSTECH
QIFK2022_88	FINE Samples for Learning with Noisy Labels	Taehyeon Kim, Jongwoo Ko	KAIST
QIFK2022_90	FIFO: Learning Fog-invariant Features for Foggy Scene Segmentation	Sohyun-Lee	POSTECH
QIFK2022_91	Time Is MattEr: Temporal Self-supervision for Video Transformers	Sukmin Yun	KAIST
QIFK2022_92	WildNet: Learning Domain Generalized Semantic Segmentation from the Wild	Suhyeon Lee	Yonsei University
QIFK2022_96	ReSTR: Convolution-free Referring Image Segmentation Using Transformers	Namyup-Kim	POSTECH
QIFK2022_103	Combating Label Distribution Shift for Active Domain Adaptation	Sehyun Hwang	POSTECH
QIFK2022_105	From Noisy Prediction to True Label: Noisy Prediction Calibration via Generative Model	HeeSun Bae, Seungjae Shin	KAIST
QIFK2022_108	Landmark-Guided Subgoal Generation in Hierarchical Reinforcement Learning	Junsu Kim	KAIST
QIFK2022_111	PartGlot: Learning Shape Part Segmentation from Language Reference Games	Juil Koo	KAIST
QIFK2022_114	Learning Semantic Segmentation from Multiple Datasets with Label Shifts	Dongwan Kim	Seoul National University
QIFK2022_118	MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentation	Inkyu Shin	KAIST
QIFK2022_121	MULTI-TASK NEURAL PROCESSES	Donggyun-Kim	KAIST
QIFK2022_123	Transformers Generalize DeepSets and Can be Extended to Graphs and Hypergraphs	Jinwoo Kim	KAIST
QIFK2022_130	CLIP-Actor: Text-Driven Recommendation and Stylization for Animating Human Meshes	Kim Youwang, Kim Ji-Yeon	POSTECH
QIFK2022_131	AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot Network	Wooseok-Lee	Seoul National University
QIFK2022_132	HandOccNet: Occlusion-Robust 3D Hand Mesh Estimation Network	JoonKyu-Park	Seoul National University
QIFK2022_138	Style Neophile: Constantly Seeking Novel Styles for Domain Generalization	Juwon-Kang	POSTECH
QIFK2022_144	Self-Calibrating Neural Radiance Fields	Yoonwoo-Jeong	POSTECH

2022 India Finalists

University	Innovation Title	Students
Indian Institute of Science, Bangalore	Reed-Muller and Polar Codes for Channels With Memory	V. Arvind Rameshwar, Nikhil Verma
Indian Institute of Science, Bangalore	Demystifying the Optimal Generator in Generative Adversarial Networks	Siddarth Asokan
Indian Institute of Science, Bangalore	Hierarchical Self-Supervised Representation Learning for Completely Blind Visual Quality Assessment	Nithin C. Babu
Indian Institute of Science, Bangalore	Deep - GAP: Robust Reconstruction for Compressed Sensing with Generative Priors	Kartheek Kumar Reddy Nareddy
Indian Institute of Science, Bangalore	Simplicial Neural Networks: A tool for Higher-order Graph Representation Learning	Sravanthi Gurugubelli
Indian Institute of Science, Bangalore	Scalable Multi-View Subspace Clustering using Sketching	Kadambari Sai Kiran
Indian Institute of Science, Bangalore	Specificity-Invariance Modulation for Source-Free Domain Adaptation	Sunandini Sanyal
Indian Institute of Technology, Bombay	Intelligent Visual Interfaces for Content Creation in Virtual Reality	Sukanya Bhattacharjee
Indian Institute of Technology, Bombay	Pure Exploration over Bit-Constrained Channels	Fathima Zarin Faizal
Indian Institute of Technology, Bombay	Representation Learning for Scalable and Multimodal Graph Retrieval	Indradyumna Roy
Indian Institute of Technology, Bombay	mm-Wave Wideband Transmitter Design for 5G NR Applications	Ajinkya Kharalkar, Anik Batabyal
Indian Institute of Technology, Delhi	Online learning for age-of-information minimization in time-varying energy harvesting networks	Akanksha Jaiswal
Indian Institute of Technology, Delhi	Sensing-Control-Communication Co-Design for 5G-V2X Triggered External Airbags in Vehicles	Mohammad Taha Shah
Indian Institute of Technology, Delhi	Lifelong learning for Neural-powered Mixed Integer Programming	Sahil Manchanda
Indian Institute of Technology, Delhi	Few-Shot Labeled Graph Generative Modeling	Sahil Manchanda, Shubham Gupta
Indian Institute of Technology, Delhi	Looking Beyond Occlusions in Dynamic LiDAR Scans	Prashant Kumar
Indian Institute of Technology, Guwahati	DEVELOPMENT OF GaN-BASED HIGHLY EFFICIENT WIDE-BANDWIDTH POWER SUPPLY	Anandh N, Somnath Meikap
Indian Institute of Technology, Guwahati	Quantification of Information Leakage and Security Verification of Compiler Optimizations	Priyanka Panigrahi
Indian Institute of Technology, Kanpur	Orthogonal Time-Frequency Space (OTFS) Modulation for Joint Radar and Communication (RadCom) in mmWave Massive MIMO Systems	Meesam Jafri, Suraj Srivastava
Indian Institute of Technology, Kanpur	Efficient Virtual Memory Management under Oversubscription Scenarios for UVM-capable GPUs	Nikhil Molugu
Indian Institute of Technology, Kanpur	Secure Estimation In 5G mmWave IOT Networks	Priyanka Maity, Kunwar Pritiraj Rajput
Indian Institute of Technology, Kharagpur	A Robust Life-long Learning Framework for Aggregation of Knowledge in Sparse Ensemble of Federated Deep Neural Networks	Ashraf Rashid
Indian Institute of Technology, Kharagpur	A Low-complexity Quantum Simulation Framework for Efficient Spectrum Estimation and Detection with Observation Noise and Quantum Uncertainties in Large Antennas	Mostafizur Rahaman Laskar
Indian Institute of Technology, Madras	StaRPlat: A Domain-Specific Language for Cross-Platform Static and Dynamic Graph Analytics	Nibedita Behera, Ashwina Kumar
Indian Institute of Technology, Madras	A High Frequency Area and Power Efficient DC-DC Converter Using Mixed Mode Timed-Based PID Controller	Abirmoya Santra
Indian Institute of Technology, Madras	Digital Signal Processing for 1.75 times Capacity Scaling of Installed Fiber Links	Sameer Ahmad MIR
Indian Institute of Technology, Madras	Enabling night-time ADAS using low-cost RGB-Thermal camera pair	Mukhesh Pugalendhi Sudha, Neham Jain
Indraprastha Institute of Information Technology, Delhi	Software/Hardware Prototype of IEEE 802.11ad/ay Based Joint Radar-Communication Transceiver	Akanksha Sneh, Aakanksha Tewari
Indraprastha Institute of Information Technology, Delhi	Inductive Representation Learning on Graphs	Karan Goyal
Indraprastha Institute of Information Technology, Delhi	Approximate Computing for prevention of Side-Channel Attacks	Tarun Sharma
Indraprastha Institute of Information Technology, Delhi	Model-Free Data Driven Estimation and Control Over A Constrained Communications Network	Shivangi Agarwal
Indraprastha Institute of Information Technology, Delhi	NERF Optimized Aerial Reconstruction	aradhya neeraj mathur
International Institute of Information Technology, Hyderabad	Artificial Intelligence Driven Self-Adaptive Low Power CMOS 28nm SerDes Transmitter Design	Abhinav Vajrala, Chanamolu Neeraj
International Institute of Information Technology, Hyderabad	List Decoding Reed-Solomon and Locally Repairable Codes: How far can we go?	Harshithanjani Athi, Rasagna Chigullapally
International Institute of Information Technology, Hyderabad	EDMOS: Emotion Driven MOvie Summarizer	Aditya Kumar Singh, Dhruv Srivastava
International Institute of Information Technology, Hyderabad	Monocular Template-free Textured 3D Digitization of Garments in-the-wild	Shanthika Naik, Chandradeep Pokhariya

2022 North America Finalists

University	Innovation Title	Students
CMU	Incentivizing the Participation of Data-Heterogeneous and Resource-Constrained Clients in Federated Learning	Yae Jee Cho, Divyansh Jhunjhunwala
CMU	Near and Far-Field Sensor Fusion for 3D Body Pose Estimation and Tracking	Yu-Jhe Li, Zhengyi Luo
CMU	Safety-Critical Scenarios Generation and Generalization for Autonomous Driving	Wenhao Ding, Jiacheng Zhu
CMU, MIT	Holistic Distributed Deep Learning Compilation with Automated Cross-stack Optimization	Byungsoo Jeon, Sunghyun Kim
Cornell	Power Inference with Self-Supervised Learning	Chenhui Deng, Andrew Butt
Georgia Tech	Enabling Privacy Preserving Computation at the Cloud using Energy Efficient and Side-Channel Secure Fully Homomorphic Encryption Accelerator for Mobile Devices	Anupam Golder, Rakshith Saligram
Georgia Tech	Moving from Black Box models to self introspecting Perception Engine for Autonomous Vehicles	Hemant Kumawat, Biswadeep Chakraborty
Georgia Tech	Secure and Privacy Aware ML Accelerator on Edge	Sudarshan Sharma, Jianming Tong
Georgia Tech, UT Austin	Real-time Visual Processing for Autonomous Driving via Video Transformer with Data-Model-Accelerator Tri-design	Rishov Sarkar, Zhiwen Fan
Michigan	Algorithm-Hardware Co-Design for Energy-Efficient Autonomous Planning	Anthony Opipari, Alphonsus Adu-Bredu
Michigan	Automatically Detecting and Preventing Hardware Prefetching Attacks	Ian Neal, Tanvir Ahmed Khan
MIT	A Data-driven Approach Towards Understanding And Synthesizing Interactions In Multi-agent Driving	Tsun-Hsuan Wang, Peter Werner
MIT	A Hardware Accelerator to Enable Computing on Encrypted Data	Nikola Samardzic, Axel Feldmann
MIT	Gallium Nitride CMOS Technology for the Next Generation of RF Front End Modules	Qingyun Xie, John Niroula
MIT	Incremental, Compositional Scene Understanding with Energy-Based Models	Yilun Du, Shuang Li
MIT	Learning Neuro-Symbolic Models for Continual Learning	Nishanth Kumar, Willie McClinton
MIT	System-Algorithm Co-Design for Efficient On-Device Training	Ji Lin, Ligeng Zhu
Princeton	Bridging the Gap between Domain-Specific Languages and Hardware Accelerators with Unified Software/Hardware Abstractions	Yu Zeng, Yi Li
Princeton	Data Multiplexing Neural Networks	Vishvak Murahari, Carlos Jimenez
Princeton	Deep Learning Enabled Inverse Design of Multi-functional mmWave PAs supporting Concurrent Multiband Operation	Emir Karahan, Zheng Liu
UCB	Backpropagation on Non-Differentiable Structures: A Fresh Approach to Discrete Structured Learning	Jonathan Shafer, Abhishek Shetty
UCB	High-Speed SerDes with Feed-Forward Non-Linear Equalizer on Silicon Photonics for Immersive AR/VR Hardware	Kunmo Kim, Sunjin Choi
UCLA	An End-to-End Full-Stack Generation Framework for Domain-Specific Accelerators	Sihao Liu, Jian Weng
UCLA	Automated Architecture Optimization using Neural Networks	Atefeh Sohrabizadeh, Yunsheng Bai
UCSB, Stanford	Ultra-Low Bandwidth Streaming of Talking-Head Videos via Generative Models	Yimeng Liu, Pulkit Tandon
UCSD	Active Learning for Efficient Annotations of Very Large Datasets: Framework, Models, Metrics, and Case Studies in Autonomous Driving	Ross Greer, Kuan-Lin Chen
UCSD	Augmenting environment sensing with multi-beam 5G mmWave systems	Ish Kumar Jain, Raghav Subbaraman
UCSD	Bridging Semantic, Geometric and Physical Reasoning in Indoor 3D Scenes	Yu-Ying Yeh, Rui Zhu
UCSD	Efficient Anomaly Detection in Sensor Networks Using Hyperdimensional Computing	Anthony Thomas, Behnam Khaleghi
UCSD	Enabling higher-order, high precision, high-speed VCO-based ADCs	Corentin Pochet, Asish Koruprolu
UCSD	Generalizing Object Pose and Shape Estimation for Robotic Manipulation	Yang Fu, Yuzhe Qin
UCSD	Neural-Enhanced Multipath-Aided Positioning	Mingchao Liang, Wenyu Zhang
UCSD	Robust Machine Learning in IoT Devices	Xiyuan Zhang, Ranak Roy Chowdhury
UCSD	Transformers as Meta-Learners for Neural Fields	Yinbo Chen, Jiarui Xu
UIUC	Certifiably Robust Sensor-fusion Systems for Autonomous Driving	Linyi Li, Zhuolin Yang
UIUC	Provably Robust Machine Learning for Wireless Systems	Zikun Liu, Calvin Xu
University of Montreal	On-Device Reinforcement Learning for Continuous Control	Pierluca D'Oro, Simone Totaro
USC	Algorithm-Hardware Co-Design for Securing the Chip and the Data on the Chip	Yuke Zhang, Dake Chen
USC	Enabling Energy-Efficient Processing-in-Pixel for Resource-constrained On-device Edge Intelligence	Gourav Datta, Zihan Yin
UT Austin	Integrated Surface Acoustic Wave-Based Fingerprint Smart Sensor Microsystem	Sinwoo Cho, Linran Zhao
UT Austin	Many-domain Processors: Securing 100s of Concurrent Security Domains	Shijia Wei, Prateek Sahu
UT Austin	Tunable ultra-small monolithically-rolled-up filter network by piezoelectric actuation	Zhendong Yang, Kristen Nguyen
Virginia Tech	Joint radar- communication based wireless localization	Gaurav Duggal, Charles Thornton
Virginia Tech	Multipath Component Identification using Reconfigurable Intelligent Surfaces (RIS): Theory and Applications to Joint Communication and Localization	Don-Roberts Emenonye, Anish Pradhan
Washington	Fully Digital Optimal Beamforming for wideband mm-Wave communication with run-time transmitter tracking and nulling of moving interferers.	Diego Pena-Colaiocco, Jung-Jin Park
Washington	Towards Next Generation Wi-Fi: An Intelligent Scheduler Design	Hao Yin, Sachin Nayak

2022 Europe Finalists

University	Title	Name	Supervisor
CISPA	Unifying Deep Generators for Privacy-preserving Data Sharing	Dingfan Chen	Mario Fritz
CISPA	Improving Advanced Digital Signatures With Applications to Electronic Payments	Benedikt Wagner	Julian Loss, Jiaxin Pan
ETH Zurich	Rowhammer Meets AI: Leveraging Deep Learning for Building a General Rowhammer Testing Methodology	Patrick Jattke	Kaveh Razavi
ETH Zurich	Principled Insights into Parameterization and Optimization Landscape of Neural Networks via the Hessian	Sidak Pal Singh	Bernhard Schölkopf
Imperial College London	Optimal Transport Theory for Generalizability and Fairness in Supervised and Imitation Learning	Sarah Boufelja	Robert Shorten
Max Planck Institute	Establishing human-interpretable DNNs as the new default	Moritz Böhle	Bernt Schiele, Mario Fritz
Tubingen	Discovering Emerging Topological Properties in Out-of-Distribution Capable Visual Similarity Systems	Karsten Roth	Zeynep Akata, Oriol Vinyals
University of Cambridge	A Unifying Framework for Sampling, Inference and Transport via Schrödinger Bridges	Francisco Vargas	Neil Lawrence, Ryan Coterell
University of Cambridge	Uncertainty-Aware Collaborative Filtering with Graph Neural Networks	Shreyas Padhy	Jose Miguel Hernandez-Lobato
University of Cambridge	Systematic Design of Neural Process Models for Probabilistic Meta-Learning	Stratis Markou	Richard Turner, Carl E. Rasmussen
University of Oxford	Rethinking Generalization in Computer Vision for the Open World	Sagar Vaze	Andrea Vedaldi, Kai Han
Vrije Universiteit Amsterdam	A Comprehensive and Scalable Mitigation Strategy for Transient Execution Vulnerabilities	Brian Johannesmeyer	Herbert Bos