Many warn that Artificial Intelligence has a serious reproducibility crisis, but is it so? Some conclusions from the author's experience trying to replicate 255 papers. | Continue reading
Anything that looks like genuine understanding is just an illusion. | Continue reading
Every day we hear claims that AI is about to transform the economy, destroying jobs & creating monopolies. But what do the professional economists think about this? | Continue reading
At 13,000 attendees, 1,428 accepted papers, and 57 workshops, NeuRIPS has unquestionably become a huge conference. Does that make it worse or better as an academic conference? A bit of both, according to one attendee. | Continue reading
Maybe every paper abstract should have a mandatory field of what the limitations of the proposed approach are. That way some of the science miscommunications and hypes could maybe be avoided.— Sebastian Risi (@risi1979) October 28, 2019 The media is often tempted to report … | Continue reading
The AI enthusiast's Introduction to Artificial Life: old ties between AI and ALife, and what makes ALife research special. | Continue reading
Humanity’s rich history has left behind an enormous number of historical documents and artifacts. However, virtually none of these documents, containing stories and recorded experiences essential to our cultural heritage, can be understood by non-experts due to language and writ … | Continue reading
I recall always having this vague impression about Gaussian Processes (GPs) being a magical algorithm that is able to define probability distributions over sets of functions, but I had always procrastinated reading up on the details. It's not completely my fault though! Whenever … | Continue reading
On different metrics for evaluating language models, the relationships among them, mathematical and empirical bounds for those metrics, and suggested best practices with regards to how to report them. | Continue reading
Since deep learning regained prominence in 2012, many machine learning frameworks have clamored to become the new favorite among researchers and industry practitioners. From the early academic outputs Caffe and Theano to the massive industry-backed PyTorch and TensorFlow, this de … | Continue reading
Progress in the field of Natural Language Processing (NLP) depends on the existence of language resources: | Continue reading
A review of Timothy Niven and Hung-Yu Kao, 2019: Probing Neural Network Comprehension of Natural Language Arguments | Continue reading
A Summary of the RSS 2019 Conference. | Continue reading
Researchers get credit for writing papers. If you’re a professor, the number of accepted papers determines whether you’ll get tenure. If you’re a student, it determines if and when you can graduate, as well as your future industry or academic job prospects. A paper is meant to | Continue reading
This June, our research team at the University of Washington released Grover, a state-of-the-art detector of neural fake news. Neural Fake News is the threat of AI-generated news articles controlled by human adversaries with the intent to deceive. Grover generates fake news … | Continue reading
AI art has a long history that is often overlooked | Continue reading
Publishing a paper in academia is challenging, stimulating, and a bit baffling. Challenging because the research might fail. Stimulating because research may start assuming one outcome and finish with a totally different one. Baffling because after the paper is written and ready, … | Continue reading
Recent advances in neural networks have generated considerable excitement about AI. But AI is not all about neural networks. Other avenues in AI research tackle problems such as building effective models of the world or logical reasoning and are especially useful for dealing with … | Continue reading
With GANs, there is a certain unpredictability that inspires, unblocks, and creates something special - something that goes far beyond Instagram filters or ordinary style transfer. | Continue reading
Imagine you're building a self-driving car that needs to understand its surroundings. How would you enable your car to perceive pedestrians, bikers, and other vehicles around it in order to move safely? You could use a camera for this, but that doesn't seem particularly effective … | Continue reading
Better use of inductive biases, human-like common sense, and unseen distributions and tasks. | Continue reading
What do robotics researchers, do? Get a glimpse in this summary of highlights from the 2018 Robotics Science and Systems Conference. | Continue reading
On the subject of private and ethical AI, the U.S. government has been disinterested, lacking in expertise, and impotent to stand up to tech corporations. | Continue reading
Learning how to learn, as we'll see, is just what we need to move beyond pure RL and leverage prior experience. | Continue reading
By definition, learning from scratch is just about the least sample-efficient approach there can be. | Continue reading
The time is ripe for practical transfer learning to make inroads into NLP. | Continue reading
In January 2017, four world-class poker players engaged in a three-week battle of heads-up no-limit Texas hold ’em. They were not competing against each other. Instead, they were fighting against a common foe: an AI system called Libratus that was developed by Carnegie Mellon res … | Continue reading
Formal theories are necessary if we want to enjoy the benefits of algorithms without the drawbacks of algorithmic bias. | Continue reading
Carlini and Wagner's latest paper opens up new possibilities in fooling speech recognition algorithms. | Continue reading
Two dangerous visions According to Engadget, 2017 was the year society started taking algorithmic bias seriously. If it’s really true—well, better late than never. Researchers have been trying to warn us for years about the dangers of putting algorithms in socially imp … | Continue reading