Learning to solve sequential tasks with recurrent models requires the ability to memorize long sequences and to extract task-relevant features from them. In this paper, we study the memorization subtask from the point of view of the design and training of recurrent neural networks. We propose a new model, the Linear Memory Network, which features an encoding-based memorization component built with a linear autoencoder for sequences. We extend the memorization component with a modular memory that encodes the hidden state sequence at different sampling frequencies. Additionally, we provide a specialized training algorithm that initializes the memory to efficiently encode the hidden activations of the network. The experimental results on synthetic and real-world datasets show that specializing the training algorithm to train the memorization component always improves the final performance whenever the memorization of long sequences is necessary to solve the problem.

We present Meena, a multi-turn open-domain chatbot trained end-to-end on data mined and filtered from public domain social media conversations. This 2.6B parameter neural network is simply trained to minimize perplexity of the next token. We also propose a human evaluation metric called Sensibleness and Specificity Average (SSA), which captures key elements of a human-like multi-turn conversation. Our experiments show strong correlation between perplexity and SSA. The fact that the best perplexity end-to-end trained Meena scores high on SSA (72% on multi-turn evaluation) suggests that a human-level SSA of 86% is potentially within reach if we can better optimize perplexity. Additionally, the full version of Meena (with a filtering mechanism and tuned decoding) scores 79% SSA, 23% higher in absolute SSA than the existing chatbots we evaluated.

The 2016 United States presidential election was marked by the abuse of targeted advertising on Facebook. Concerned with the risk of the same kind of abuse to happen in the 2018 Brazilian elections, we designed and deployed an independent auditing system to monitor political ads on Facebook in Brazil. To do that we first adapted a browser plugin to gather ads from the timeline of volunteers using Facebook. We managed to convince more than 2000 volunteers to help our project and install our tool. Then, we use a Convolution Neural Network (CNN) to detect political Facebook ads using word embeddings. To evaluate our approach, we manually label a data collection of 10k ads as political or non-political and then we provide an in-depth evaluation of proposed approach for identifying political ads by comparing it with classic supervised machine learning methods. Finally, we deployed a real system that shows the ads identified as related to politics. We noticed that not all political ads we detected were present in the Facebook Ad Library for political ads. Our results emphasize the importance of enforcement mechanisms for declaring political ads and the need for independent auditing platforms.

You may be dazzled, spooked or annoyed by the ability of online retailers to predict which products you're interested in purchasing--but what if you could play the same game? Whether it's Amazon recommending soap or Netflix suggesting a movie, more companies are making personalized predictions using variations on a machine learning tactic called market basket analysis. This technique uses an algorithm that sorts through behavioral data to determine how frequently certain actions (purchases, views, etc.) are associated with other actions. The algorithm provides the statistical likelihood that if one action takes place, another desired action is likely to follow. Sophisticated algorithms aren't limited to shopping carts or movie recommendations, however.

It took me an embarrassingly long time to realize that the "black mirror" of the popular anthology series Black Mirror was a screen, or rather, all the screens we surround ourselves with: phones, tablets, computers, TVs, and, increasingly, futuristic devices built by massive corporations that monitor our movements and preferences and words. We buy these black mirrors, welcoming them into our homes and lives and letting them -- true to their name -- reflect ourselves back to us. And as we know all too well, those reflections sometimes betray our darkest impulses. Unsettling reflections are not the black mirrors' fault. Gadgets are merely assemblages of wires and metal and glass. Devices don't have a point of view; they operate according to the input they receive, the algorithms and designs and patterns that power the software, written by humans and thus shaded and slanted by human biases.

Sign in to report inappropriate content. We hear a lot about artificial intelligence, especially on social media. Watch this episode to learn more!

That is what SynctacticAI is: a platform-as-a-service, dedicated to AI and Machine learning.