십분딥러닝_18_GumBolt (VAE with Boltzmann Machine)
Download as pptx, pdf0 likes261 views
GumBolt (VAE with Boltzmann Machine) 에 관한 설명입니다.
1 of 15
Downloaded 67 times
Ad
Recommended
[RAG Tutorial] 02. RAG 프로젝트 파이프라인.pdf
[RAG Tutorial] 02. RAG 프로젝트 파이프라인.pdfHyunKyu Jeon[RAG Tutorial] RAG 프로젝트 파이프라인
[GITHUB] https://github.com/hkjeon13/rag-tutorial
[RAG Tutorial] 01. RAG 프로젝트 개요.pdf
[RAG Tutorial] 01. RAG 프로젝트 개요.pdfHyunKyu Jeon[RAG Tutorial] 프로젝트 개요
[GITHUB] https://github.com/hkjeon13/rag-tutorial
[PR-358] Training Differentially Private Generative Models with Sinkhorn Dive...
[PR-358] Training Differentially Private Generative Models with Sinkhorn Dive...HyunKyu Jeon[PR-358] Paper Review - Training Differentially Private Generative Models with Sinkhorn Divergence
Super tickets in pre trained language models
Super tickets in pre trained language modelsHyunKyu JeonThis document discusses finding "super tickets" in pre-trained language models through pruning attention heads and feedforward layers. It shows that lightly pruning BERT models can improve generalization without degrading accuracy (phase transition phenomenon). The authors propose a new pruning approach for multi-task fine-tuning of language models called "ticket sharing" where pruned weights are shared across tasks. Experiments on GLUE benchmarks show their proposed super ticket and ticket sharing methods consistently outperform unpruned baselines, with more significant gains on smaller tasks. Analysis indicates pruning reduces model variance and some tasks share more task-specific knowledge than others.
Synthesizer rethinking self-attention for transformer models
Synthesizer rethinking self-attention for transformer models HyunKyu JeonSynthesizer: Rethinking Self-Attention for Transformer Models
Domain Invariant Representation Learning with Domain Density Transformations
Domain Invariant Representation Learning with Domain Density TransformationsHyunKyu JeonThe document discusses domain invariant representation learning aimed at creating models that generalize well to unseen domains, contrasting it with domain adaptation. It proposes a method that enforces invariance across transformations between domains and utilizes generative adversarial networks to implement these transformations. The effectiveness of the proposed approach is demonstrated on various datasets, achieving competitive results compared to state-of-the-art methods in domain generalization.
Meta back translation
Meta back translationHyunKyu JeonThis document summarizes Meta Back-Translation, a method for improving back-translation by training the backward model to directly optimize the performance of the forward model during training. The key points are:
1. Back-translation typically relies on a fixed backward model, which can lead the forward model to overfit to its outputs. Meta back-translation instead continually trains the backward model to generate pseudo-parallel data that improves the forward model.
2. Experiments show Meta back-translation generates translations with fewer pathological outputs like greatly differing in length from references. It also avoids both overfitting and underfitting of the forward model by flexibly controlling the diversity of pseudo-parallel data.
3. Related work leverages mon
Maxmin qlearning controlling the estimation bias of qlearning
Maxmin qlearning controlling the estimation bias of qlearningHyunKyu JeonThis document summarizes the Maxmin Q-learning paper published at ICLR 2020. Maxmin Q-learning aims to address the overestimation bias of Q-learning and underestimation bias of Double Q-learning by maintaining multiple Q-functions and using the minimum value across them for the target in the Q-learning update. It defines the action selection and target construction for the update based on taking the maximum over the minimum Q-value for each action. The algorithm initializes multiple Q-functions, selects a random subset to update using the maxmin target constructed from the minimum Q-values. This approach reduces the biases seen in prior methods.
Adversarial Attack in Neural Machine Translation
Adversarial Attack in Neural Machine TranslationHyunKyu JeonAdversarial Attack in Neural Machine Translation
십분딥러닝_15_SSD(Single Shot Multibox Detector)
십분딥러닝_15_SSD(Single Shot Multibox Detector)HyunKyu JeonSSD(Single Shot Multibox Detector)에 대한 설명입니다.
십분딥러닝_13_Transformer Networks (Self Attention)
십분딥러닝_13_Transformer Networks (Self Attention)HyunKyu Jeon셀프어텐션(Self Attention)기법을 사용한 Transformer Networks에 대한 설명입니다.
십분수학상관괶계(ǰپDz)
십분수학상관괶계(ǰپDz)HyunKyu Jeon상관관계(Correlation), 피어슨 상관계수 (Pearson Correlation Coefficient), 스피어만 순위 상관계수(Spearman Rank Correlation Coefficient), 켄달 순위 상관계수(Kendall Rank Correlation Coefficient)
십분수학거리(پٲԳ)
십분수학거리(پٲԳ)HyunKyu JeonDistance in terms of Data Sciences/Math
about Euclidean Distance, Manhattan Distance, Minkowski Distance, and Mahalanobis distance.
More Related Content
More from HyunKyu Jeon (20)
Adversarial Attack in Neural Machine Translation
Adversarial Attack in Neural Machine TranslationHyunKyu JeonAdversarial Attack in Neural Machine Translation
십분딥러닝_15_SSD(Single Shot Multibox Detector)
십분딥러닝_15_SSD(Single Shot Multibox Detector)HyunKyu JeonSSD(Single Shot Multibox Detector)에 대한 설명입니다.
십분딥러닝_13_Transformer Networks (Self Attention)
십분딥러닝_13_Transformer Networks (Self Attention)HyunKyu Jeon셀프어텐션(Self Attention)기법을 사용한 Transformer Networks에 대한 설명입니다.
십분수학상관괶계(ǰپDz)
십분수학상관괶계(ǰپDz)HyunKyu Jeon상관관계(Correlation), 피어슨 상관계수 (Pearson Correlation Coefficient), 스피어만 순위 상관계수(Spearman Rank Correlation Coefficient), 켄달 순위 상관계수(Kendall Rank Correlation Coefficient)
십분수학거리(پٲԳ)
십분수학거리(پٲԳ)HyunKyu JeonDistance in terms of Data Sciences/Math
about Euclidean Distance, Manhattan Distance, Minkowski Distance, and Mahalanobis distance.
Adversarial Attack in Neural Machine Translation
Adversarial Attack in Neural Machine TranslationHyunKyu Jeon
십분딥러닝_13_Transformer Networks (Self Attention)
십분딥러닝_13_Transformer Networks (Self Attention)HyunKyu Jeon