Machine Learning

For this week (1/3/2021–7/3/2021), I will be reading the following 2 research papers.Photo by freddie marriage on UnsplashAuthors: Miryam ElizabethVilla-Péreza, Miguel Á.Álvarez-Carmonab, OctavioLoyola-Gonzálezc, Miguel Angel Medina-Péreza, Juan Carlos Velazco-Rossellc and Kim-Kwang Raymond ChoodVenue: Knowledge-Based SystemsPaper: URLAbstract:While anomaly detection is … [Read more...] about Weekly Machine Learning Research Paper Reading List — #12

저희 AITRICS에서는 딥러닝 모델을 기반으로 패혈증을 조기에 예측하는 의료기기 소프트웨어 VitalCare를 만들었습니다. VitalCare에 적용되는 인공지능 모델 개발의 단계는 크게 ‘데이터 확보-데이터 전처리 및 레이블 작업-모델 학습-모델 평가-임상시험을 통한 모델 성능 확인’ 5단계로 나누어 볼 수 있습니다.먼저 첫 단계인 ‘데이터 확보’는 딥러닝 모델 개발에 있어 가장 필수적인 단계라고 말씀드릴 수 있습니다. 성능이 좋은 딥러닝 모델을 만들기 위해서는 최신 모델 구조들과 학습 알고리즘을 적용하는 것도 중요하게 작용할 수 있지만, 그보다도 우선적으로 양질의 데이터를 최대한 많이 확보하는 것이 가장 중요합니다. 이를 위해 AITRICS는 가장 대표적으로 사용되는 … [Read more...] about [VitalCare] #2 기존 평가도구의 한계를 극복하는 인공지능

Meaning is the building block of human language, uncovered by context.Meaning is core to language because the meaning of a sentence determines the forms of words and phrases that are selected and vice versa. Or as I say: Form follows meaning®. But what is meaning?In language, the word forms that we use to communicate with others follow the meaning of what we want to say and, … [Read more...] about Using Meaning as Universal Knowledge Representation

Photo credit: Plaid.comIt is critical that both businesses and customers have confidence in the Reepei platform. One way to build confidence is to increase both parties’ understanding of how the platform works, specifically repayment. Let’s look into how the platform predicts this important number.The context surrounding inability to payAt Reepei, we believe that debts must be … [Read more...] about How Reepei Predicts Your Repayment Amount

Yes, but hear me out!Credit: PixabayDeep learning has become ubiquitous with data science and an inextricable element of machine learning. It’s shaped how humans interact with machines perhaps more so than any other advance in mathematical modeling to date. With self-driving cars, computers triumphing over masters at their respective table games, and language translation … [Read more...] about An Attack on Deep Learning

There are two main considerations when it comes to adopting TFX: value and cost. I want to demonstrate the value of TFX and how it helps with production-level experimentation, and adopting the best standards for model and data validation.The cost for using TFX is all about locking up Machine Learning to Tensorflow stacks, such as model training and data transformation. Moving … [Read more...] about Why we need TensorFlow Extended (TFX) in 3 diagrams

Explaining a model using concepts from game theorySource (Unsplash)In an increasing number of domains, machine learning models have started being held to a higher standard. Model predictions are no longer enough. Companies can now be held responsible for any spurious predictions their models produce. With this shift, model explainability has arguably taken priority over … [Read more...] about SHAP Values for Model Interpretation

The below shows the code for a signal function that detects divergences. The function is very simple and further improvements can be made to make it smarter but at the moment, it is sufficient enough to detect what the eye sees.def divergence(Data, indicator, lower_barrier, upper_barrier, width, buy, sell):for i in range(len(Data)):try:if Data[i, indicator] < … [Read more...] about The Percent Bollinger Indicator. Creating & Coding a Trading Strategy.

Veri ön işleme, model kurulmadan önce veri seti üzerinde yapılan bir takım düzeltmeler, eksik gözlemlerin çeşitli yöntemlerle giderilmesi, verileri kaldırma, dönüştürme gibi işlemlerdir. Özetle, veriyi sindirmek ve makine öğrenmesi modellerinin veriden daha fazla ve daha doğru bir şekilde yararlanmasını sağlamak için kullanılan teknik/yöntemler bütününe Veri Ön İşleme … [Read more...] about Veri Ön İşleme — 1

After loading BERT large model for Q&A when you print the model it will show all the layers and sub layers of the model. BERT large has 24 bert layer, 1 BertEmbedding layer at beginning and qa_output layer at the end.In this blog I will talk about sections in the Embedding layer and not the bert layer as I am assuming that readers already know about self attention , … [Read more...] about Implementing BERT for Question and Answer