The ability to convey relevant and diverse information is critical in multi-document summarization and yet remains elusive for neural seq-to-seq models whose outputs are often redundant and fail to correctly cover important details. In this work, we propose an attention mechanism which encourages greater focus on relevance and diversity. Attention weights are computed based on (proportional) probabilities given by Determinantal Point Processes (DPPs) defined on the set of content units to be summarized. DPPs have been successfully used in extractive summarisation, here we use them to select relevant and diverse content for neural abstractive summarisation. We integrate DPP-based attention with various seq-to-seq architectures ranging from CNNs to LSTMs, and Transformers. Experimental evaluation shows that our attention mechanism consistently improves summarization and delivers performance comparable with the state-of-the-art on the MultiNews dataset.

Autoregressive models are now capable of generating high-quality minute-long expressive MIDI piano performances. Even though this progress suggests new tools to assist music composition, we observe that generative algorithms are still not widely used by artists due to the limited control they offer, prohibitive inference times or the lack of integration within musicians' workflows. In this work, we present the Piano Inpainting Application (PIA), a generative model focused on inpainting piano performances, as we believe that this elementary operation (restoring missing parts of a piano performance) encourages human-machine interaction and opens up new ways to approach music composition. Our approach relies on an encoder-decoder Linear Transformer architecture trained on a novel representation for MIDI piano performances termed Structured MIDI Encoding. By uncovering an interesting synergy between Linear Transformers and our inpainting task, we are able to efficiently inpaint contiguous regions of a piano performance, which makes our model suitable for interactive and responsive A.I.-assisted composition. Finally, we introduce our freely-available Ableton Live PIA plugin, which allows musicians to smoothly generate or modify any MIDI clip using PIA within a widely-used professional Digital Audio Workstation.

Modern keyboards allow a musician to play multiple instruments at the same time by assigning zones -- fixed pitch ranges of the keyboard -- to different instruments. In this paper, we aim to further extend this idea and examine the feasibility of automatic instrumentation -- dynamically assigning instruments to notes in solo music during performance. In addition to the online, real-time-capable setting for performative use cases, automatic instrumentation can also find applications in assistive composing tools in an offline setting. Due to the lack of paired data of original solo music and their full arrangements, we approach automatic instrumentation by learning to separate parts (e.g., voices, instruments and tracks) from their mixture in symbolic multitrack music, assuming that the mixture is to be played on a keyboard. We frame the task of part separation as a sequential multi-class classification problem and adopt machine learning to map sequences of notes into sequences of part labels. To examine the effectiveness of our proposed models, we conduct a comprehensive empirical evaluation over four diverse datasets of different genres and ensembles -- Bach chorales, string quartets, game music and pop music. Our experiments show that the proposed models outperform various baselines. We also demonstrate the potential for our proposed models to produce alternative convincing instrumentations for an existing arrangement by separating its mixture into parts. All source code and audio samples can be found at https://salu133445.github.io/arranger/ .

AI and machine learning is data-driven, Lange said, and now we train computers instead of programming them.

If you've been having trouble finding love on dating apps, you might want to try dating one of your friends, a new study suggests. The study authors, based in British Columbia, Canada looked at data from just under 2,000 couples of different demographics. They found two thirds started out as just friends, suggesting that establishing a platonic connection with someone first is conducive to a solid romantic relationship later. The study suggests that the cliché of falling in love at first site – a frequent trope in the Hollywood movies of the silver screen – is slightly outdated in the 21st century. Built on a more solid foundation?

I don't think it's a spoiler to say that the machines are trying to take over in the Netflix show The Mitchells vs. the Machines. I mean, there's obviously some type of conflict. But in case you haven't seen it yet, here is your official warning: I'm going to use the movie to do some fun estimation problems about the robot apocalypse. Maybe I've already said too much. Let's get right to the important stuff: The machines have decided that they would be better off without all those pesky humans, so they are gathering up all the people and putting them into seven giant 128-story rockets.

In a recent demo conducted over Zoom, I watched as the music composition app Dynascore transformed the entire emotional tenor of a short video multiple times in under a minute, all without altering a single frame of the visuals. What began in my short briefing as a very serious workout ad with a very serious soundtrack--something where you'd expect to see neon sweat pouring out of the athlete's head behind a Gatorade logo--quickly changed in tone to something a bit funnier. The machine intelligence engine inside Dynascore swapped out the action film theme music for Beethoven's somber Moonlight Sonata, suddenly transforming the video into a dark comedy. A few taps of the mouse on the other end of the Zoom window, a few seconds of rendering, and I was watching the same video with a modern pop song now layered over it, equally form-fitting to the burly close-ups on screen. This time, the result felt more like a music video.

Roku has been slowly expanding its line of affordable soundbars, most recently with last year's compact, "pretty good" Streambar, and now comes the $180 Streambar Pro, the successor to 2019's Smart Soundbar. While none of Roku's soundbars will appeal to audiophiles, they're perfect for everyday viewers looking to upgrade their TV's tinny audio without breaking the bank, and the 2.0-channel Streambar Pro ups the ante with surprisingly solid sound and a new virtual surround mode. The Streambar Pro's efficient audio performance is already a strong selling point, but the real draw is an integrated Roku streaming player with 4K HDR playback, not to mention Alexa, Google Assistant, HomeKit, and AirPlay 2 support. This review is part of TechHive's coverage of the best soundbars. Click that link to read reviews of competing products, along with a buyer's guide to the features you should consider when shopping.

In this hands-on project, we will train a Bidirectional Neural Network and LSTM based deep learning model to detect fake news from a given news corpus. This project could be practically used by any media company to automatically predict whether the circulating news is fake or not. The process could be done automatically without having humans manually review thousands of news related articles. This project is for anyone with foundation in programming and machine learning who wants to develop Data science and Machine learning projects but having limited resources on their computer and limited time. You will learn how to use the Google Colaboratory via your web browser to develop a Fake and Real News Detection Data Science Project.

It is because of this that astronomers are turning towards machine learning and Artificial Intelligence (AI) in order to build new tools.