The accounts are likely just parodies, not part of a sinister corporate strategy, but they illustrate the kind of thing that could happen someday.
The news: Ahead of a landmark vote that could lead to the formation of the first-ever labor union at a US-based Amazon warehouse, new Twitter accounts purporting to be Amazon employees started appearing. The profiles used deepfake photos as profile pictures and were tweeting some pretty laughable, over-the-top defenses of Amazon’s working practices. They didn’t seem real, but they still led to confusion among the public. Was Amazon really behind them? Was this some terrible new anti-union social media strategy? The answer is almost certainly not—but the use of deepfakes in this context points to a more concerning trend overall. Read More
Monthly Archives: April 2021
Why A.I. knows who you find attractive better than you do
When it comes to earning social currency, being attractive is as good as gold.
A team of scientists from Finland has now designed a machine learning algorithm that can plumb the depths of these subjective judgments better than we can and can accurately predict who we find attractive via our unique brainwaves — and even generate a unique portrait that captures these qualities — with 83 percent accuracy.
Far beyond just the laws of attraction, this novel brain-computer interface (BCI) could push wide-open a new era of BCI that can bring our unvoiced desires to life.
The research was published this February in the journal IEEE Transactions on Affective Computing. Read More
Lost Tapes of the 27 Club
Using AI to create the album lost to music’s mental health crisis.
As long as there’s been popular music, musicians and crews have struggled with mental health at a rate far exceeding the general adult population. And this issue hasn’t just been ignored. It’s been romanticized, by things like the 27 Club—a group of musicians whose lives were all lost at just 27 years old.
To show the world what’s been lost to this mental health crisis, we’ve used artificial intelligence to create the album the 27 Club never had the chance to. Through this album, we’re encouraging more music industry insiders to get the mental health support they need, so they can continue making the music we all love for years to come.
Because even AI will never replace the real thing. Read More
If Your Company Uses AI, It Needs an Institutional Review Board
Conversations around AI and ethics may have started as a preoccupation of activists and academics, but now — prompted by the increasing frequency of headlines of biased algorithms, black box models, and privacy violations — boards, C-suites, and data and AI leaders have realized it’s an issue for which they need a strategic approach.
A solution is hiding in plain sight. Other industries have already found ways to deal with complex ethical quandaries quickly, effectively, and in a way that can be easily replicated. Instead of trying to reinvent this process, companies need to adopt and customize one of health care’s greatest inventions: the Institutional Review Board, or IRB. Read More
Ensemble Learning: Bagging & Boosting
How to combine weak learners to build a stronger learner to reduce bias and variance in your ML model
The bias and variance tradeoff is one of the key concerns when working with machine learning algorithms. Fortunately there are some Ensemble Learning based techniques that machine learning practitioners can take advantage of in order to tackle the bias and variance tradeoff, these techniques are bagging and boosting. So, in this blog we are going to explain how bagging and boosting works, what theirs components are and how you can implement them in your ML problem. Read More
Google starts trialing its FLoC cookie alternative in Chrome
Google today announced that it is rolling out Federated Learning of Cohorts (FLoC), a crucial part of its Privacy Sandbox project for Chrome, as a developer origin trial.
FLoC is meant to be an alternative to the kind of cookies that advertising technology companies use today to track you across the web. Instead of a personally identifiable cookie, FLoC runs locally and analyzes your browsing behavior to group you into a cohort of like-minded people with similar interests (and doesn’t share your browsing history with Google). That cohort is specific enough to allow advertisers to do their thing and show you relevant ads, but without being so specific as to allow marketers to identify you personally.
This “interest-based advertising,” as Google likes to call it, allows you to hide within the crowd of users with similar interests. All the browser displays is a cohort ID and all your browsing history and other data stay locally. Read More
Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks
We identify label errors in the test sets of 10 of the most commonly-used computer vision, natural language, and audio datasets, and subsequently study the potential for these label errors to affect benchmark results. Errors in test sets are numerous and widespread: we estimate an average of 3.4% errors across the 10 datasets,2where for example 2916 label errors comprise 6% of the ImageNet validation set.Putative label errors are identified using confident learning algorithms and then human-validated via crowdsourcing (54% of the algorithmically-flagged candidates are indeed erroneously labeled). Traditionally, machine learning practitioners choose which model to deploy based on test accuracy — our findings advise caution here, proposing that judging models over correctly labeled test sets maybe more useful, especially for noisy real-world datasets. Surprisingly, we find that lower capacity models may be practically more useful than higher capacity models in real-world datasets with high proportions of erroneously labeled data.For example, on ImageNet with corrected labels: ResNet-18 outperforms ResNet-50 if the prevalence of originally mislabeled test examples increases by just 6%. OnCIFAR-10 with corrected labels: VGG-11 outperforms VGG-19 if the prevalence of originally mislabeled test examples increases by just 5%. Read More
#accuracy, #biasTransfer Learning and Data Augmentation applied to the Simpsons Image Dataset
Deep Learning application using Tensorflow and Keras
In the ideal scenario for Machine Learning (ML), there are abundant labeled training instances, which share the same distribution as the test data [1]. However, these data can be resource-intensive or unrealistic to collect in certain scenarios. Thus, Transfer Learning (TL) becomes a useful approach. It consists of increasing the learning ability of a model by transferring information from a different but related domain. In other words, it relaxes the hypothesis that the training and testing data are independent and identically distributed [2]. It only works if the features that are intended to be learned are general to both tasks. Another method to work with limited data is by using Data Augmentation (DA). It consists of applying a suite of transformations to inflate the dataset. Traditional ML algorithms rely significantly on feature engineering, while Deep Learning (DL) focuses on learning data by unsupervised or semi-supervised feature learning methods and hierarchical feature extraction. DL often requires massive amounts of data to be trained effectively, making it a strong candidate for TL and DA. Read More
I Dream My Painting and I Paint My Dream
Dutch photographer Bas Uterwijk used artificial intelligence to create a realistic portrait of Vincent van Gogh on van Gogh’s 168th birthday.
Rick's Cafe AI 