창업가들이 알아야할 지식재산 전략을 사례와 이론을 중심으로 작성하였습니다.
특히, 성공하고 있는 BLT의 고객사례를 많이 다루었습니다.
- 특허권 이론 및 전략
- 정부지원사업을 통한 초기 창업전략
- 디자인권과 저작권의 비교
- 상표이론 및 좋은 브랜드 네이밍
문의 : shawn@blte.kr (엄정한 변리사)
The document discusses various machine learning clustering algorithms like K-means clustering, DBSCAN, and EM clustering. It also discusses neural network architectures like LSTM, bi-LSTM, and convolutional neural networks. Finally, it presents results from evaluating different chatbot models on various metrics like validation score.
The document discusses challenges with using reinforcement learning for robotics. While simulations allow fast training of agents, there is often a "reality gap" when transferring learning to real robots. Other approaches like imitation learning and self-supervised learning can be safer alternatives that don't require trial-and-error. To better apply reinforcement learning, robots may need model-based approaches that learn forward models of the world, as well as techniques like active localization that allow robots to gather targeted information through interactive perception. Closing the reality gap will require finding ways to better match simulations to reality or allow robots to learn from real-world experiences.
[243] Deep Learning to help student’s Deep Learning
This document describes research on using deep learning to predict student performance in massive open online courses (MOOCs). It introduces GritNet, a model that takes raw student activity data as input and predicts outcomes like course graduation without feature engineering. GritNet outperforms baselines by more than 5% in predicting graduation. The document also describes how GritNet can be adapted in an unsupervised way to new courses using pseudo-labels, improving predictions in the first few weeks. Overall, GritNet is presented as the state-of-the-art for student prediction and can be transferred across courses without labels.
[234]Fast & Accurate Data Annotation Pipeline for AI applications
This document provides a summary of new datasets and papers related to computer vision tasks including object detection, image matting, person pose estimation, pedestrian detection, and person instance segmentation. A total of 8 papers and their associated datasets are listed with brief descriptions of the core contributions or techniques developed in each.
This document presents a formula for calculating the loss function J(θ) in machine learning models. The formula averages the negative log likelihood of the predicted probabilities being correct over all samples S, and includes a regularization term λ that penalizes predicted embeddings being dissimilar from actual embeddings. It also defines the cosine similarity term used in the regularization.
창업가들이 알아야할 지식재산 전략을 사례와 이론을 중심으로 작성하였습니다.
특히, 성공하고 있는 BLT의 고객사례를 많이 다루었습니다.
- 특허권 이론 및 전략
- 정부지원사업을 통한 초기 창업전략
- 디자인권과 저작권의 비교
- 상표이론 및 좋은 브랜드 네이밍
문의 : shawn@blte.kr (엄정한 변리사)
The document discusses various machine learning clustering algorithms like K-means clustering, DBSCAN, and EM clustering. It also discusses neural network architectures like LSTM, bi-LSTM, and convolutional neural networks. Finally, it presents results from evaluating different chatbot models on various metrics like validation score.
The document discusses challenges with using reinforcement learning for robotics. While simulations allow fast training of agents, there is often a "reality gap" when transferring learning to real robots. Other approaches like imitation learning and self-supervised learning can be safer alternatives that don't require trial-and-error. To better apply reinforcement learning, robots may need model-based approaches that learn forward models of the world, as well as techniques like active localization that allow robots to gather targeted information through interactive perception. Closing the reality gap will require finding ways to better match simulations to reality or allow robots to learn from real-world experiences.
[243] Deep Learning to help student’s Deep LearningNAVER D2
This document describes research on using deep learning to predict student performance in massive open online courses (MOOCs). It introduces GritNet, a model that takes raw student activity data as input and predicts outcomes like course graduation without feature engineering. GritNet outperforms baselines by more than 5% in predicting graduation. The document also describes how GritNet can be adapted in an unsupervised way to new courses using pseudo-labels, improving predictions in the first few weeks. Overall, GritNet is presented as the state-of-the-art for student prediction and can be transferred across courses without labels.
[234]Fast & Accurate Data Annotation Pipeline for AI applicationsNAVER D2
This document provides a summary of new datasets and papers related to computer vision tasks including object detection, image matting, person pose estimation, pedestrian detection, and person instance segmentation. A total of 8 papers and their associated datasets are listed with brief descriptions of the core contributions or techniques developed in each.
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지NAVER D2
This document presents a formula for calculating the loss function J(θ) in machine learning models. The formula averages the negative log likelihood of the predicted probabilities being correct over all samples S, and includes a regularization term λ that penalizes predicted embeddings being dissimilar from actual embeddings. It also defines the cosine similarity term used in the regularization.
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기NAVER D2
The document discusses running a TensorFlow Serving (TFS) container using Docker. It shows commands to:
1. Pull the TFS Docker image from a repository
2. Define a script to configure and run the TFS container, specifying the model path, name, and port mapping
3. Run the script to start the TFS container exposing port 13377
The document discusses linear algebra concepts including:
- Representing a system of linear equations as a matrix equation Ax = b where A is a coefficient matrix, x is a vector of unknowns, and b is a vector of constants.
- Solving for the vector x that satisfies the matrix equation using linear algebra techniques such as row reduction.
- Examples of matrix equations and their component vectors are shown.
This document describes the steps to convert a TensorFlow model to a TensorRT engine for inference. It includes steps to parse the model, optimize it, generate a runtime engine, serialize and deserialize the engine, as well as perform inference using the engine. It also provides code snippets for a PReLU plugin implementation in C++.
The document discusses machine reading comprehension (MRC) techniques for question answering (QA) systems, comparing search-based and natural language processing (NLP)-based approaches. It covers key milestones in the development of extractive QA models using NLP, from early sentence-level models to current state-of-the-art techniques like cross-attention, self-attention, and transfer learning. It notes the speed and scalability benefits of combining search and reading methods for QA.
1. Music is Social. 레알?
(SN을 활용한 싸이BGM추천)
2011.10.18
CTO / 기술연구소 / Data Science Team. 박태수
This report contains information that is confidential and proprietary to SK Communications and is solely for the use of SK Communications personnel.
No part of it may be used, circulated, quoted, or reproduced for distribution outside SK Communications. If you are not the intended recipient of this
report, you are hereby notified that the use, circulation, quoting, or reproducing of this report is strictly prohibited and may be unlawful.
SK Communications
17. Music is Social . 레알?
Music is Social. 레알?
일촌 네트워크 사이에서는
뚜렷한 경향성을 보이지 않는다.
SK Communications
18. Music is Social . 레알?
Social 정보를 어떻게 활용할까?
• Social Music Chart
일촌 Chart(Borda count)
• Social Music Recommendation
Collaborative Filtering
SK Communications
19. Music is Social . 레알?
Social Music Chart
Social Chart(Borda Count)
• Borda Rank Normalization
U : a set of items
τ : rank
• Borda Score
SK Communications
20. Music is Social . 레알?
Social Music Chart
Social Chart(Borda Count)
• 일촌들의 BGM 재생 횟수에 따른 Ranking
BGM 일촌1 일촌2 일촌3 일촌4 • Borda Rank Normalization
Hello-허각 1 1 2
안녕이라고 말하지마-다비치 2 2 4
U : a set of items
Who Am I - 유브이 3 1 1
τ : rank
In Heaven - JYJ 4 3 3
… … … … …
• Borda Count를 적용한 Social Music Chart
BGM 일촌1 일촌2 일촌3 일촌4 … SUM Rank
Hello-허각 1-((1-1)/10)=1 0.5+((2-1)/(2*10))=0.55 1-((1-1)/10)=1 1-((2-1)/10)=0.9 … 3.45 1
안녕이라고 말하지마 -다비치 1-((2-1)/10)=0.9 1-((2-1)/10)=0.9 0.5+((3-1)/(2*10))=0.6 1-((4-1)/10)=0.7 … 3.1 3
Who Am I - 유브이 1-((3-1)/10)=0.8 1-((1-1)/10)=1 0.5+((3-1)/(2*10))=0.6 1-((1-1)/10)=1 … 3.4 2
In Heaven - JYJ 1-((4-1)/10)=0.7 0.5+((2-1)/(2*10))=0.55 1-((3-1)/10)=0.8 1-((3-1)/10)=0.8 … 2.85 4
SK Communications
21. Music is Social . 레알?
Collaborative Filtering
많은 사용자들로부터 얻은 기호정보에 따라 사용자들의 관심사를 자동으로 예측하게 해주는 방법
User1
A B C D
Nearest Recommend
Neighbor User2
A B C
User3
V F D E
SK Communications
22. Music is Social . 레알?
Collaborative Filtering
Collaborative Filtering의 종류
사용자 기반 필터링
사용자 간의 유사성을 측정하여 선호도가 비슷한 다른 고객들이 평가한
상품을 기반으로 특정 고객이 선호할만한 상품을 추천하는 방식
데이터 양이 작고, 데이터 변경이 자주 일어나는 경우
항목 기반 필터링
고객이 선호도 등급을 입력한 기존 상품들과 추천하고자 하는 상품들 간의
유사성을 측정하여 특정 고객이 어떤 상품을 선호하는지 예측하여 추천
데이터 양이 크고, 데이터 변경이 자주 일어나지 않는 경우
SK Communications
23. Music is Social . 레알?
Collaborative Filtering
Item Similarity Computation
Pearson correlation coefficient
SK Communications
24. Music is Social . 레알?
Collaborative Filtering
Prediction Computation
Pu ,: 고객 u의 아이템 i에 대한 선호도 예측값
i
Pu ,i
allSimilla rItems, N
( S i , N * Ru , N )
R : 고객 u의 아이템 N에 대한 Rating
u,N
allSimilarItems, N
(| S i , N |)
S : 상품 i와 다른 상품 N의 유사도 값
i, N
Item 유사도 Matrix 선호도 예측
- 1과 5 : 0.9
- 2와 5 : 0.8
- 3과 5 : 0.2
- 4와 5 : 0.1
BGM1 BGM2 BGM3 BGM4 BGM5
User1 5 4 1 2 ?
User2 1 2 5 5 1
SK Communications
25. Music is Social . 레알?
Collaborative Filtering
음악 서비스를 처음 사용하는 사용자라면??
구매, 재생정보가 없다면???
사용자의 지인들의 Social Music Chart를 이용
SK Communications
26. Music is Social . 레알?
맺음말
Music is social. 레알?
Social Network 활용 방안
Social Music Chart
Social Music Recommendation
SK Communications
27. Music is Social . 레알?
Q&A
Thank you for your attention!
SK Communications