An associative memory is a framework of content-addressable memory that stores a collection of message vectors (or a dataset) over a neural network while enabling a neurally feasible mechanism to recover any message in the dataset from its noisy version. Designing an associative memory requires addressing two main tasks: 1) learning phase: given a dataset, learn a concise representation of the dataset in the form of a graphical model (or a neural network), 2) recall phase: given a noisy version of a message vector from the dataset, output the correct message vector via a neurally feasible algorithm over the network learnt during the learning phase. This paper studies the problem of designing a class of neural associative memories which learns a network representation for a large dataset that ensures correction against a large number of adversarial errors during the recall phase. Specifically, the associative memories designed in this paper can store dataset containing $\exp(n)$ $n$-length message vectors over a network with $O(n)$ nodes and can tolerate $\Omega(\frac{n}{{\rm polylog} n})$ adversarial errors. This paper carries out this memory design by mapping the learning phase and recall phase to the tasks of dictionary learning with a square dictionary and iterative error correction in an expander code, respectively.

Researchers are developing an app which could help to prevent suicides by flagging those most at risk. Using a computer algorithm, it records conversations, analysing what people say and how they speak. By picking up on a range of subtle verbal and non-verbal cues, it can correctly classify if someone is suicidal with 93 per cent accuracy. At the heart of the app is a machine learning algorithm which classifies the person based on their responses. In an earlier study, researchers enrolled a mix of 379 patients, who were suicidal, diagnosed as mentally ill, or neither.

A robot has just set a new record for the fastest-solved Rubik's Cube, according to its makers. The Sub1 Reloaded robot took just 0.637 seconds to analyse the toy and make 21 moves, so that each of the cube's sides showed a single colour. That beats a previous record of 0.887 seconds, which was achieved by an earlier version of the same machine using a different processor. Infineon provided its chip to highlight advancements in self-driving car tech. But one expert has questioned the point of the stunt.

A new speed record has been set for solving the notoriously difficult Rubik's Cube. TC Newman (@PurpleTCNewman) has the details. A link has been sent to your friend's email address. A new speed record has been set for solving the notoriously difficult Rubik's Cube. TC Newman (@PurpleTCNewman) has the details.

The Rubik's cube was devised by Hungarian architect Erno Rubik more than 30 years ago, but he likely never envisioned his puzzle being cracked this quickly. A robot has this week solved a Rubik's cube in 0.637 seconds, at the Electronica Trade Fair in Munich, Germany. The machine, known as'Sub1 Reloaded' and developed by German tech company Infineon, was aided by one of the world's most powerful microcomputers. The machine, known as'Sub1 Reloaded' and developed by German tech company Infineon, was aided by one of the world's most powerful microcomputers The robot took a fraction of a second to analyse the cube and make 21 moves to solve the puzzle. Its time of 0.637 seconds beat the previous world record of 0.887 seconds, set by an earlier prototype of the same machine.

The shape of multicoloured three-dimensional puzzle Rubik's Cube is not a trademark, the European Court of Justice has ruled. It means the shape of the cube alone is not enough to protect it from being copied. UK company Seven Towns, which manages Rubik's Cube's intellectual property rights, registered its shape as a trademark in in the 1990s. But German firm Simba Toys challenged the trademark protection in 2006. The European Court of Justice (ECJ) agreed that the cube's ability to rotate should be protected by a patent and not a trademark.

It's the familiar multicoloured "cube" brain-teaser which has challenged puzzle solvers for more than 40 years and is still the world's bestselling toy of all time. But on Thursday – following a 10-year legal tussle – Rubik's Cube lost a key trademark battle after the European court of justice (ECJ) said its shape was not sufficient to grant it protection from "copycat" versions. The eponymous puzzle, invented in 1974 by Hungarian sculptor and architect Erno Rubik, is popular among young and old, with more than 350m cubes sold to date worldwide. UK company Seven Towers, which oversees Rubik's Cube intellectual property rights, registered the shape as a three-dimensional EU trademark with the European Union Intellectual Property Office (EUIPO) in April 1999. But the court ruled that the EU trademark representing the shape of the Rubik's Cube is invalid, triggering fears it will lead to a surge of cheap, mass-produced versions and a weakening of European intellectual property protection.

The easy answer is that excel is a resource hog & despite being a tremendously powerful tool for mining small datasets, when you start to push past the traditional 65565 rows, you start to move into realms where Microsoft is traditionally not good (memory handling, i/o management, efficient processing). First a few questions: 1. Cardinality: Were both tables internally unique? Excel bogs down on cartesian products in my experience -- you need one to many or one to one matches. Am assuming you ran a pivot table on both counting the unique instances of each email address, comparing the row count in the table with the grand total (both should be the same). Am assuming that this answer is going to be yes across the board, since you're probably a really good excel jockey.

The dream of thinking machines goes back centuries, at least to Gottfried Wilhelm Leibniz, in the 17th century. Leibniz (right) helped invent mechanical calculators, independently of Isaac Newton developed the integral calculus, and had a lifelong fascination with reducing thinking to calculation. His Mathesis Universalis was a vision of universal science made possible by a mathematical language more precise than natural languages, like English. The Limits of Modern AI: A Story In the 18th Century the Enlightenment philosopher and proto-psychologist Étienne Bonnot de Condillac imagined a statue outwardly appearing like a man and also with what he called "the inward organization." In an example of supreme armchair speculation, Condillac imagined pouring facts--bits of knowledge--into its head, wondering when intelligence would emerge. Condillac's musings drew inspiration from the early mechanical philosophy of Thomas Hobbes, who had famously declared that thinking was nothing but ...

Nuance Communications Inc. announced that the Dragon Drive connected car framework will now have contextual reasoning which will allow AI to deliver contextualized and personalized recommendations for navigation, dining, gas stations, points of interest (POI) and other domains. The service can be integrated across a variety of Dragon Drive's Domain applications designed specifically for the in-car experience, and is entirely accessible through a conversational user interface or haptic feedback where appropriate. The Contextual Reasoning Framework leverages Nuance's advancements in Artificial Intelligence to exploit domain knowledge in a context-sensitive manner to provide a more intelligent experience behind the wheel. A core element of Nuance's Automotive Assistant offering, the new Contextual Reasoning Framework can proactively provide information and make recommendations by performing reasoning on a large repository of knowledge, in combination with a driver's preferences, location and situational context, and the data from the car's sensors. Each entry of the knowledge repository can range over general, spatial and temporal knowledge, domain-specific knowledge about parking, fuel stations, and information specific to certain geographic markets and vehicle characteristics.