• Adversarial Attack 1
• Automatic Speech Recognition 4
• Compression 1
• Deep Learning 4
• Develop 3
• Essay 1
• Generalization 1
• Korean 3
• Machine Learning 1
• Memory 6
• Model Serving 1
• NLP 9
• Pretrained LM 1
• Pruning 1
• Quantization 2
• Recommendation 1
• Reinforce Learning 1
• Research 1
• Tensorflow 2 4
• Transducer 1
• Transformer 3
• huggingface 1

Adversarial Attack

• DOING MORE WITH LESS: IMPROVING ROBUSTNESS USING GENERATED DATA 리뷰

Automatic Speech Recognition

• Tensorflow2 기반 ASR 모델, 학습 코드 구현
• Conformer: Convolution-augmented Transformer for Speech Recognition 리뷰
• Sequence Transduction with Recurrent Neural Networks 리뷰
• Listen, Attend and Spell 리뷰

Compression

• Training with Quantization Noise for Extreme Model Compression 리뷰

Deep Learning

• Neural Turing Machines 리뷰
• Exploring Generalization in Deep Learning 리뷰
• What Do Compressed Deep Neural Networks Forget? 리뷰
• Training with Quantization Noise for Extreme Model Compression 리뷰

Develop

• Tensorflow2 기반 ASR 모델, 학습 코드 구현
• Tensorflow2 기반 Seq2Seq 모델, 학습, 서빙 코드 구현
• 한국어 띄어쓰기 교정 모델 개발 (Quickspacer)

Essay

• Research Vision

Generalization

• Exploring Generalization in Deep Learning 리뷰

Korean

• AI X Bookathon 4회 후기
• huggingface를 이용한 한국어 BART 학습 후기
• 한국어 띄어쓰기 교정 모델 개발 (Quickspacer)

Machine Learning

• Exploring Generalization in Deep Learning 리뷰

Memory

• ∞-former: Infinite Memory Transformer 요약
• Memformer: A Memory-Augmented Transformer for Sequence Modeling 리뷰
• Compressive Transformers for Long-Range Sequence Modelling 리뷰
• Sequential Recommendation with User Memory Networks 리뷰
• Not all memories are created equal: Learning to forget by expiring 리뷰
• Mem2Seq: Effectively Incorporating Knowledge Bases into End-to-End Task-Oriented Dialog Systems 요약

Model Serving

• TorchServe 사용법과 후기

NLP

• AI X Bookathon 4회 후기
• Not all memories are created equal: Learning to forget by expiring 리뷰
• Mem2Seq: Effectively Incorporating Knowledge Bases into End-to-End Task-Oriented Dialog Systems 요약
• huggingface를 이용한 한국어 BART 학습 후기
• Tensorflow2 기반 Seq2Seq 모델, 학습, 서빙 코드 구현
• BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension 리뷰
• What Do Compressed Deep Neural Networks Forget? 리뷰
• Training with Quantization Noise for Extreme Model Compression 리뷰
• 한국어 띄어쓰기 교정 모델 개발 (Quickspacer)

Pretrained LM

• BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension 리뷰

Pruning

• What Do Compressed Deep Neural Networks Forget? 리뷰

Quantization

• What Do Compressed Deep Neural Networks Forget? 리뷰
• Training with Quantization Noise for Extreme Model Compression 리뷰

Recommendation

• Sequential Recommendation with User Memory Networks 리뷰

Reinforce Learning

• Decision Transformer: Reinforcement Learning via Sequence Modeling 요약

Research

• Research Vision

Tensorflow 2

• huggingface를 이용한 한국어 BART 학습 후기
• Tensorflow2 기반 ASR 모델, 학습 코드 구현
• Tensorflow2 기반 Seq2Seq 모델, 학습, 서빙 코드 구현
• 한국어 띄어쓰기 교정 모델 개발 (Quickspacer)

Transducer

• Sequence Transduction with Recurrent Neural Networks 리뷰

Transformer

• ∞-former: Infinite Memory Transformer 요약
• Memformer: A Memory-Augmented Transformer for Sequence Modeling 리뷰
• Compressive Transformers for Long-Range Sequence Modelling 리뷰

huggingface

• huggingface를 이용한 한국어 BART 학습 후기