If you are looking for an answer to the question What is Artificial Intelligence? and you only have a minute, then here's the definition the Association for the Advancement of Artificial Intelligence offers on its home page: "the scientific understanding of the mechanisms underlying thought and intelligent behavior and their embodiment in machines."
However, if you are fortunate enough to have more than a minute, then please get ready to embark upon an exciting journey exploring AI (but beware, it could last a lifetime) …
Reprinted with permission from Quanta Magazine's Abstractions blog. It's very easy to break things in biology," said Loren Frank, a neuroscientist at the University of California, San Francisco. "It's really hard to make them work better." Yet against the odds, researchers at the New York University School of Medicine reported earlier this summer that they had improved the memory of lab animals by tinkering with the length of a dynamic signal in their brains--a signal that has fascinated neuroscientists like Frank for decades. The feat is exciting in its own right, with the potential to enhance recall in people someday, too. But it also points to a more comprehensive way of thinking about memory, and it identifies an important clue, rooted in the duration of a neural event, that could pave the way to a greater understanding of how memory works. Since the 1980s, scientists have been tuning in to short bursts of synchronized neural activity in the brain area called the hippocampus.
Synthetic aperture sonar (SAS) imagery can generate high resolution images of the seafloor. Thus, segmentation algorithms can be used to partition the images into different seafloor environments. In this paper, we compare two possibilistic segmentation approaches. Possibilistic approaches allow for the ability to detect novel or outlier environments as well as well known classes. The Possibilistic Fuzzy Local Information C-Means (PFLICM) algorithm has been previously applied to segment SAS imagery. Additionally, the Possibilistic K-Nearest Neighbors (PKNN) algorithm has been used in other domains such as landmine detection and hyperspectral imagery. In this paper, we compare the segmentation performance of a semi-supervised approach using PFLICM and a supervised method using Possibilistic K-NN. We include final segmentation results on multiple SAS images and a quantitative assessment of each algorithm.
How do we remember our experiences? The mental skill of bringing previously encountered people, events, and objects to mind is intuitive, but how neural circuits enable this episodic memory retrieval remains a fundamental question in neuroscience. On page 975 of this issue, Vaz et al. (1) use intracranial electrophysiological recordings in humans to identify a putative mechanism involved in memory retrieval: synchronized occurrence of high-frequency oscillations across brain regions. Their findings highlight the importance of dynamic interactions between brain areas in mediating complex cognitive processes and suggest a biomarker for pinpointing neural populations involved in different memories.
A liquid state machine (LSM) is a particular kind of spiking neural network. An LSM consists of a large collection of units (called nodes, or neurons). Each node receives time varying input from external sources (the inputs) as well as from other nodes. Nodes are randomly connected to each other. The recurrent nature of the connections turns the time varying input into a spatio-temporal pattern of activations in the network nodes.
Why it matters: Earlier this month, we reported that the impact of artificial intelligence is happening only slowly, but will pick up to a rapid clip in the late 2020s. But with automation, we are already there, according to a report by the Geneva-based World Economic Forum. The findings align generally with a slew of reports released the last two years by Western think tanks and companies. "Mass redeployment in labor forces around the world, including reskilling/retraining (rather than mass unemployment), is one of the societal grand challenges in the next several decades," said Michael Chui, who has led numerous automation studies for the McKinsey Global Institute. In the U.S., there is only embryonic discussion underway.
Cryptocurrency and artificial intelligence technologies took center stage this week at Slush Tokyo, one of the nation's largest tech conferences and an annual event that brings together startups, investors and more established businesses. This year's conference featured many speakers, panel discussions and startups moving into the rapidly growing world of cryptocurrency and AI. Cryptocurrencies, or virtual currencies, can usually be traded without the intermediation of a central authority, with the most famous being bitcoin. But the underlying technology has spawned a number of other similar virtual currencies and other new funding mechanisms. "This is the fourth time we had the event," said Antti Sonninen, CEO of Slush Tokyo.
Bitcoin's price has dropped by 6.15 per cent over the last 24 hours - while almost all of the alternatives to the market leading cryptocurrency are also in decline. At the time of publication, all but two of the top 50 digital currencies by market capitalisation have fallen in value since yesterday, according to CoinMarketCap. Bitcoin is now valued at $10,526 (£7,594) as a result of the drop. Ethereum, the second most valuable digital currency, is down 7.62 per cent to $799.61 (£576.96) Bitcoin cash and litecoin, which complete the top five, are down 6.81 per cent and 7.01 per cent respectively.
Looking back, 2017 was a significant year for technology. Consider that at the beginning of 2017, Bitcoin was virtually unknown to most people, and Artificial Intelligence was something that they had perhaps heard of once or twice, but was still a concept confined to science fiction. Cloud computing was still untested to many people, and hardly anyone would realise how prevalent Bots would become a mere 12 months later.
Not a day goes by when one is not bombarded by the latest innovations around artificial intelligence (AI), robotics and machine learning (ML). The inflationary use of these terms makes many people question if they are simply catchy buzzwords ― part of a short-lived market hype. On the other hand, expectations concerning the capabilities of AI and robotics are at an all-time high. From the ultimate AI-built utopia to Skynet apocalypse ― everything seems possible. Time for a grounded look at what AI, ML and robotics actually can and should do in the area of finance process automation.