Contests & Prizes


Trump intervenes to let Afghan teens attend robot competition in US

FOX News

U.S. officials have reportedly decided to allow a group of Afghan girls, who has previously been denied entry, into the U.S. to attend a robot competition. Following the change, the State Department said it worked with the Department of Homeland Security to remedy the situation. "The State Department worked incredibly well with the Department of Homeland Security to ensure that this case was reviewed and handled appropriately," Dina Powell, White House deputy national security adviser for strategy, said in a statement, according to Politico. I look forward to welcoming this brilliant team of Afghan girls, and their competitors, to Washington DC next week!


The good, the bad and the ugly: Artificial intelligence Access AI

#artificialintelligence

The Defense Advanced Research Projects Agency (DARPA) was so interested in the potential impact of AI on computer security that it was willing to place a £44M bet to find out what that potential was. In this, seven teams were granted access to a DARPA-supplied super computer and tasked with writing a bot that could independently seek out and identify software vulnerabilities, write proof of concept exploit code to exploit them and at the same time, patch and defend their own system from attack. If AI and machine learning promises hackers such an invaluable tool, why aren't we seeing it widely used today? Today, software vulnerabilities remain far too plentiful and humans far too susceptible to social engineering attacks to justify the effort.


BBVA announces Change as the winner of its Open Talent Artificial Intelligence competition - BBVA NEWS

#artificialintelligence

Themes ranged from providing more personalised products and services, better risk management, offering customers greater insights into their transaction, personalised money management systems and real-time AI chatbots. Axyon AI from Italy: offers Deep Learning-powered Artificial Intelligence solutions for finance businesses like hedge funds. Sentimer from Spain: Sentimer Technologies is an Artificial Intelligence chatbot platform for customer acquisition, cross-selling and service for banking, insurance companies and financial services providers. Spin Analytics from UK: Spin Analytics brings digital transformation in Credit Risk Management by leveraging predictive analytics, AI and ML techniques on Big Data.


How Feature Engineering Can Help You Do Well in a Kaggle Competition – Part 3

@machinelearnbot

The feature interactions option was enabled, meaning that for all possible two-paired feature combinations, feature values were multiplied and hashed (read about Feature Hashing in the first post) to a position in a sparse feature vector with dimension of 2²⁸. The input data was again categorical features, with the addition of some selected numeric binned features (read about Feature Binning in the first post). A recent variant of Factorization Machines, named Field-aware Factorization Machines (FFM) was used in winning solutions for two CTR prediction competitions in 2014. This ensemble considered as input features the best 3 FFM and 3 FTRL model predictions as well as 15 selected engineered numeric features (like user views count, user preference similarity and average CTR by categories).


2nd Place Solution to 2017 DSB - Daniel Hammack and Julian de Wit

@machinelearnbot

Julian and I independently wrote summaries of our solution to the 2017 Data Science Bowl. Anyway, the LUNA16 dataset had some very crucial information - the locations in the LUNA CT scans of 1200 nodules. Data augmentation is a crucial but subtle part of my solution, and in general is one of the reasons that neural networks are so great for computer vision problems. It's the reason that I am able to build models on only 1200 samples (nodules) and have them work very well (normal computer vision datasets have 10,000 - 10,000,000 images).


How Feature Engineering Can Help You Do Well in a Kaggle Competition – Part 2

@machinelearnbot

That post described some preliminary and important data science tasks like exploratory data analysis and feature engineering performed for the competition, using a Spark cluster deployed on Google Dataproc. It was necessary to separate from clicks_train.csv Machine learning models were trained using train set data and their accuracy was evaluated on validation set data, by comparing the predictions with the ground truth labels (clicks). As we optimize CV model accuracy -- by testing different feature engineering approaches, algorithms and hyperparameters tuning -- we expect to improve our score on the competition Leaderboard (LB) accordingly (test set). The categorical fields whose average CTR presented higher predictive accuracy on CV score were ad_document_id, ad_source_id, ad_publisher_id, ad_advertiser_id, ad_campain_id, document attributes (category_ids, topics_ids, entities_ids) and their combinations with event_country_id, which modeled regional user preferences.


Recognizing Emotions using Artificial Intelligence – Produvia Blog

#artificialintelligence

Convolutional Neural Networks (CNNs) are leading the Computer Vision (CV) industry in achieving state-of-the-art facial expression (emotion) recognition. There are a few Kaggle data science competitions that you can reference for facial keypoint and expression recognition. Want to push your data science skills and deep learning skills? APIs allow you to implement artificial intelligence technologies quickly and easily.


Mind Reading: Using Artificial Neural Nets to Predict Viewed Image Categories From EEG Readings - Silicon Valley Data Science

#artificialintelligence

I chose to work on one of the datasets suggested by DeepGram: EEG readings from a Stanford research project that predicted which category of images their test subjects were viewing using linear discriminant analysis. Winning Kaggle competition teams have successfully applied artificial neural networks on EEG data (see first place winner of the grasp-and-lift challenge and third place winner of seizure prediction competition). The six major categories of images shown to test subjects were: human body, human face, animal body, animal face, natural object, and man-made object. The two plots below show the training history of the CNN model's accuracy and categorical cross entropy loss on the test data set as well as the holdout data set (labeled as "validation" in the plots).


How Feature Engineering Can Help You Do Well in a Kaggle Competition – Part 2

@machinelearnbot

That post described some preliminary and important data science tasks like exploratory data analysis and feature engineering performed for the competition, using a Spark cluster deployed on Google Dataproc. It was necessary to separate from clicks_train.csv Machine learning models were trained using train set data and their accuracy was evaluated on validation set data, by comparing the predictions with the ground truth labels (clicks). As we optimize CV model accuracy -- by testing different feature engineering approaches, algorithms and hyperparameters tuning -- we expect to improve our score on the competition Leaderboard (LB) accordingly (test set). The categorical fields whose average CTR presented higher predictive accuracy on CV score were ad_document_id, ad_source_id, ad_publisher_id, ad_advertiser_id, ad_campain_id, document attributes (category_ids, topics_ids, entities_ids) and their combinations with event_country_id, which modeled regional user preferences.


Startup Founder's Quest for Cure Leads to Genomics Hackathon at Google Xconomy

#artificialintelligence

Without any ready-made solutions on hand from big drug companies and their established research teams, Faber started to recruit individuals to his cause. Faber's grassroots initiative has led to a signal event this weekend--an AI Genomics Hackathon involving hundreds of artificial intelligence engineers and life sciences researchers, hosted by Google, which is providing $150,000 worth of processing power and a site for the mass collaboration at Google Launchpad in San Francisco. Starting Friday, June 23, the volunteer experts will combine their skills to analyze a dataset that includes a high-quality whole genome sequence of Faber's genome, and the sequence of a tumor he developed as a result of a disorder diagnosed as neurofibromatosis type 2 (NF2). Kane leads Silicon Valley Artificial Intelligence (SVAI), a self-organized community of AI enthusiasts focused on the use of machine learning and other computational tools in the life sciences.