​Deep Sensor Fusion for Real-Time Odometry Estimation

Authors: Michelle Valente, Cyril Joly, Arnaud de La Fortelle

Abstract: …consecutive frames. Results on a real road dataset show that the fusion network runs in real-time and is able to improve the odometry estimation of a single sensor alone by learning how to fuse two different types of data information.
https://arxiv.org/abs/1908.00524

🔗 Deep Sensor Fusion for Real-Time Odometry Estimation
Cameras and 2D laser scanners, in combination, are able to provide low-cost, light-weight and accurate solutions, which make their fusion well-suited for many robot navigation tasks. However, correct data fusion depends on precise calibration of the rigid body transform between the sensors. In this paper we present the first framework that makes use of Convolutional Neural Networks (CNNs) for odometry estimation fusing 2D laser scanners and mono-cameras. The use of CNNs provides the tools to not only extract the features from the two sensors, but also to fuse and match them without needing a calibration between the sensors. We transform the odometry estimation into an ordinal classification problem in order to find accurate rotation and translation values between consecutive frames. Results on a real road dataset show that the fusion network runs in real-time and is able to improve the odometry estimation of a single sensor alone by learning how to fuse two different types of data information.

​"Songs to sing in the car" - recommender systems at Spotify (human vs machine) | AI Podcast Clips

🔗 "Songs to sing in the car" - recommender systems at Spotify (human vs machine) | AI Podcast Clips
This is a clip from a conversation with Gustav Soderstrom on the Artificial Intelligence podcast. You can watch the full conversation here: https://bit.ly/2yzx6hN or watch other AI clips here: https://bit.ly/2JYkbfZ Gustav Soderstrom is the Chief Research & Development Officer at Spotify, leading Product, Design, Data, Technology & Engineering teams. Full episode: https://bit.ly/2yzx6hN Clips playlist: https://bit.ly/2JYkbfZ Full episodes playlist: https://bit.ly/2EcbaKf Podcast website: https://lexfridman.com

​Treat Negation Stopwords Differently According to Your NLP Task
The negation words (not, nor, never) are considered to be stopwords in NLTK, spacy and sklearn, but we should pay different attention
https://towardsdatascience.com/treat-negation-stopwords-differently-according-to-your-nlp-task-e5a59ab7c91f?source=collection_home---4------1-----------------------

🔗 Treat Negation Stopwords Differently According to Your NLP Task
The negation words (not, nor, never) are considered to be stopwords in NLTK, spacy and sklearn, but we should pay different attention…

​Shifts & Twists in Business Analytics: Reflections from Qlik Qonnections and Alteryx Inspire
Business Analytics evolves mostly upward, but with unexpected shifts in thinking and sudden twists in technology.
https://towardsdatascience.com/https-towardsdatascience-com-shifts-twists-in-business-analytics-bec8766091b9?source=topic_page---------2------------------1

🔗 Shifts & Twists in Business Analytics: Reflections from Qlik Qonnections and Alteryx Inspire
Business Analytics evolves mostly upward, but with unexpected shifts in thinking and sudden twists in technology.

​LSTM-based African Language Classification
Tired of German-French dataset? Look at Yemba, and stand out. Mechanics of LSTM, GRU explained and applied, with powerful visuals and code.
https://towardsdatascience.com/lstm-based-african-language-classification-e4f644c0f29e?source=collection_home---4------0-----------------------

🔗 LSTM-based African Language Classification
Tired of German-French dataset? Look at Yemba, and stand out. Mechanics of LSTM, GRU explained and applied, with powerful visuals and code.

​Deep Generative Model Driven Protein Folding Simulation

Authors: Heng Ma, Debsindhu Bhowmik, Hyungro Lee, Matteo Turilli, Michael T. Young, Shantenu Jha, Arvind Ramanathan

Abstract: …ensemble of MD runs, and (2) identifying novel states from which simulations can be initiated to sample rare events (e.g., sampling folding events).
https://arxiv.org/abs/1908.00496

🔗 Deep Generative Model Driven Protein Folding Simulation
Significant progress in computer hardware and software have enabled molecular dynamics (MD) simulations to model complex biological phenomena such as protein folding. However, enabling MD simulations to access biologically relevant timescales (e.g., beyond milliseconds) still remains challenging. These limitations include (1) quantifying which set of states have already been (sufficiently) sampled in an ensemble of MD runs, and (2) identifying novel states from which simulations can be initiated to sample rare events (e.g., sampling folding events). With the recent success of deep learning and artificial intelligence techniques in analyzing large datasets, we posit that these techniques can also be used to adaptively guide MD simulations to model such complex biological phenomena. Leveraging our recently developed unsupervised deep learning technique to cluster protein folding trajectories into partially folded intermediates, we build an iterative workflow that enables our generative model to be coupled with

​Location-Based Recommendations
Create a location-based recommender with machine learning.
https://towardsdatascience.com/location-based-recommendations-bb70af7b1538?source=collection_home---4------2-----------------------

🔗 Location-Based Recommendations
Create a location-based recommender with machine learning.

​Stanford University School of Engineering CS224N Natural Language Processing with Deep Learning

https://www.youtube.com/playlist?list=PLU40WL8Ol94IJzQtileLTqGZuXtGlLMP_

🔗 CS224N Natural Language Processing with Deep Learning - YouTube
https://web.stanford.edu/class/cs224n/syllabus.html

#data #mining #weka
Data mining with Weka | Data mining Tutorial for Beginners

🎥 Data mining with Weka | Data mining Tutorial for Beginners
👁 1 раз ⏳ 13741 сек.
In this data mining course you will learn how to do data mining tasks with Weka. This #data #mining course has been designed for beginners. It will walk you through data mining process with #weka. Weka is very powerful and widely used for data mining.

Enjoy this data mining course and share with those who need it.

Topic Covered:
***************
Data Mining with Weka (1.1: Introduction)

Data Mining with Weka (1.2: Exploring the Explorer)

Data Mining with Weka (1

​Dog Breed prediction using CNNs and transfer learning
In this article, I will demonstrate how to use keras and tensorflow to build, train, and test a Convolutiuonal Neural Network
https://towardsdatascience.com/dog-breed-prediction-using-cnns-and-transfer-learning-22d8ed0b16c5?source=collection_home---4------2-----------------------

🔗 Dog Breed prediction using CNNs and transfer learning
In this article, I will demonstrate how to use keras and tensorflow to build, train, and test a Convolutiuonal Neural Network capable of…

​Несколько соображений по поводу параллельных вычислений в R применительно к «enterprise» задачам

#BigData

Параллельные или распределенные вычисления — вещь сама по себе весьма нетривиальная. И среда разработки должна поддерживать, и DS специалист должен обладать навыками проведения параллельных вычислений, да и задача должна быть приведена к разделяемому на части виду, если таковой существует. Но при грамотном подходе можно весьма ускорить решение задачи однопоточным R, если у вас под руками есть хотя бы многоядерный процессор (а он есть сейчас почти у всех), с поправкой на теоретическую границу ускорения, определяемую законом Амдала. Однако, в ряде случаев даже его можно обойти.

https://habr.com/ru/post/462469/

🔗 Несколько соображений по поводу параллельных вычислений в R применительно к «enterprise» задачам
Параллельные или распределенные вычисления — вещь сама по себе весьма нетривиальная. И среда разработки должна поддерживать, и DS специалист должен обладать навы...

​Ученые заявили ИИ в качестве автора нового патента и пытаются изменить патентное законодательство

#Искусственныйинтеллект

Ученые и юристы из Великобритании пытаются одержать верх в споре с патентными чиновниками относительно регистрации одного изобретения. Эта команда заявила в качестве автора изобретения искусственный интеллект. Патентная заявка, о которой идет речь, подана сразу в трех разных странах.

Dabus AI, такое название получил ИИ, заявлен в качестве автора изобретения в Великобритании, ЕС и США. По словам команды, которая инициировала весь процесс, процесс регистрации патентов должен быть изменен с тем, чтобы в качестве авторов можно было заявлять не только людей. Собственно, ученые желают просто показать, что законодательство большинства стран не готово к изменениям, которые происходят уже сейчас.
https://habr.com/ru/company/madrobots/blog/462477/

🔗 Ученые заявили ИИ в качестве автора нового патента и пытаются изменить патентное законодательство
Ученые и юристы из Великобритании пытаются одержать верх в споре с патентными чиновниками относительно регистрации одного изобретения. Эта команда заявила в ка...

Машинное обучение в IOT устройствах (demo)

🎥 Машинное обучение в IOT устройствах (demo)
👁 1 раз ⏳ 601 сек.
Машинное обучение в IOT устройствах (ru)
Никита Бедункевич - Head of hardware development OneSoil
Обзор существующих решений и архитектур для встраиваемых систем. Реализация нейроускорителя на базе FPGA для IoT.

​George Hotz: Comma.ai, OpenPilot, and Autonomous Vehicles | Artificial Intelligence (AI) Podcast

🔗 George Hotz: Comma.ai, OpenPilot, and Autonomous Vehicles | Artificial Intelligence (AI) Podcast
George Hotz is the founder of Comma.ai, a machine learning based vehicle automation company. He is an outspoken personality in the field of AI and technology in general. He first gained recognition for being the first person to carrier-unlock an iPhone, and since then has done quite a few interesting things at the intersection of hardware and software. This conversation is part of the Artificial Intelligence podcast. INFO: Podcast website: https://lexfridman.com/ai YouTube Playlist: https://bit.ly/2EcbaKf

​Kaggle Etiquette | Kaggle

🔗 Kaggle Etiquette | Kaggle
There might not be any teacups or finger sandwiches on the Kaggle forums but that doesn't mean there aren't certain rules of etiquette you should probably follow. Kaggle data scientist Rachael will help you learn how to mind your p's and q's... Kaggle style! SUBSCRIBE: https://www.youtube.com/c/kaggle?sub_confirmation=1&utm_medium=youtube&utm_source=channel&utm_campaign=yt-sub About Kaggle: Kaggle is the world's largest community of data scientists. Join us to compete, collaborate, learn, and do your data

​Neural Network Compression Framework (NNCF)
This module contains a PyTorch-based framework and samples for neural networks compression. The framework organized as a Python module that can be built and used in a standalone mode. The framework architecture is unified to make it easy to add different compression methods. The samples demonstrate the usage of compression algorithms for three different use cases on public models and datasets: Image Classification, Object Detection and Semantic Segmentation.
https://github.com/opencv/openvino_training_extensions/tree/develop/pytorch_toolkit/nncf

🔗 opencv/openvino_training_extensions
Trainable models and NN optimization tools. Contribute to opencv/openvino_training_extensions development by creating an account on GitHub.

Double Deep Q Learning Is Simple with Keras

https://www.youtube.com/watch?v=UCgsv6tMReY

🎥 Double Deep Q Learning Is Simple with Keras
👁 1 раз ⏳ 2820 сек.
In this tutorial you are going to code a double deep Q learning agent in Keras, and beat the lunar lander environment. Double Q Learning resolves the inherent bias in Q learning by decoupling action selection and action-value estimation.

Silver et al. showed in 2015 that we can get significantly better results than vanilla deep Q learning, in the Atari environments.

Simple Deep Q Network w/Pytorch: https://youtu.be/UlJzzLYgYoE
Reinforcement Learning Crash Course: https://youtu.be/sOiNMW8k4T0
Policy Gradi

​How To Use Data Science For Social Impact
Transparency • Corruption • Collaboration
https://towardsdatascience.com/how-to-use-data-science-for-social-impact-e9b272b1a4b3?source=collection_home---4------2-----------------------

🔗 How To Use Data Science For Social Impact
Transparency • Corruption • Collaboration • Trust

​Federated Learning: A New AI Business Model
Federated learning is not only a promising technology but also a possible brand new AI business model. Indeed, as a consultant

https://towardsdatascience.com/federated-learning-a-new-ai-business-model-ec6b4141b1bf?source=collection_home---4------0-----------------------

🔗 Federated Learning: A New AI Business Model
Federated learning is not only a promising technology but also a possible brand new AI business model. Indeed, as a consultant, I have…

​Bayesian Modeling Airlines Customer Service Twitter Response Time
Student’s t-distribution, Poisson distribution, Negative Binomial distribution, Hierarchical modeling and Regression
https://towardsdatascience.com/bayesian-modeling-airlines-customer-service-twitter-response-time-74af893f02c0?source=collection_home---4------0-----------------------

🔗 Bayesian Modeling Airlines Customer Service Twitter Response Time
Student’s t-distribution, Poisson distribution, Negative Binomial distribution, Hierarchical modeling and Regression

​YOLO, YOLOv2 and YOLOv3: All You want to know

🔗 YOLO, YOLOv2 and YOLOv3: All You want to know
During last few years Object detection become one of the hottest areas of computer vision , and many researchers are racing to get the…