Math word problems provide a natural abstraction to a range of natural language understanding problems that involve reasoning about quantities, such as interpreting election results, news about casualties, and the financial section of a newspaper. Units associated with the quantities often provide information that is essential to support this reasoning. This paper proposes a principled way to capture and reason about units and shows how it can benefit an arithmetic word problem solver. This paper presents the concept of Unit Dependency Graphs (UDGs), which provides a compact representation of the dependencies between units of numbers mentioned in a given problem. Inducing the UDG alleviates the brittleness of the unit extraction system and allows for a natural way to leverage domain knowledge about unit compatibility, for word problem solving. We introduce a decomposed model for inducing UDGs with minimal additional annotations, and use it to augment the expressions used in the arithmetic word problem solver of (Roy and Roth 2015) via a constrained inference framework. We show that introduction of UDGs reduces the error of the solver by over 10 %, surpassing all existing systems for solving arithmetic word problems. In addition, it also makes the system more robust to adaptation to new vocabulary and equation forms .

Math word problems provide a natural abstraction to a range of natural language understanding problems that involve reasoning about quantities, such as interpreting election results, news about casualties, and the financial section of a newspaper. Units associated with the quantities often provide information that is essential to support this reasoning. This paper proposes a principled way to capture and reason about units and shows how it can benefit an arithmetic word problem solver. This paper presents the concept of Unit Dependency Graphs (UDGs), which provides a compact representation of the dependencies between units of numbers mentioned in a given problem. Inducing the UDG alleviates the brittleness of the unit extraction system and allows for a natural way to leverage domain knowledge about unit compatibility, for word problem solving. We introduce a decomposed model for inducing UDGs with minimal additional annotations, and use it to augment the expressions used in the arithmetic word problem solver of (Roy and Roth 2015) via a constrained inference framework. We show that introduction of UDGs reduces the error of the solver by over 10 %, surpassing all existing systems for solving arithmetic word problems. In addition, it also makes the system more robust to adaptation to new vocabulary and equation forms .

Math word problems provide a natural abstraction to a range of natural language understanding problems that involve reasoning about quantities, such as interpreting election results, news about casualties, and the financial section of a newspaper. Units associated with the quantities often provide information that is essential to support this reasoning. This paper proposes a principled way to capture and reason about units and shows how it can benefit an arithmetic word problem solver. This paper presents the concept of Unit Dependency Graphs (UDGs), which provides a compact representation of the dependencies between units of numbers mentioned in a given problem. Inducing the UDG alleviates the brittleness of the unit extraction system and allows for a natural way to leverage domain knowledge about unit compatibility, for word problem solving. We introduce a decomposed model for inducing UDGs with minimal additional annotations, and use it to augment the expressions used in the arithmetic word problem solver of (Roy and Roth 2015) via a constrained inference framework. We show that introduction of UDGs reduces the error of the solver by over 10 %, surpassing all existing systems for solving arithmetic word problems. In addition, it also makes the system more robust to adaptation to new vocabulary and equation forms .

MOSCOW/UNITED NATIONS – Russia released drone footage Monday showing new destruction in Syria's historic town of Palmyra, which was recently recaptured by the Islamic State group, and warned that the militants could be planning the further demolition of antiquities. The Russian Defense Ministry says Syrian government forces are advancing toward the town as another battle for the ancient site looms. The video showed that the militants have badly damaged the facade of the Roman-era amphitheater and the Tetrapylon -- a set of four monuments with four columns each at the center of the colonnaded road leading to the theater. The video appears to show that only two of the 16 columns remain standing. IS militants have destroyed ancient sites across their self-styled Islamic caliphate in Syria and Iraq, perceiving them as monuments to idolatry.

It has animal hairs coated with ionic liquid gel to pollinate a flower like a bee. They may be small, but bees and other insects play a critical role in pollination and maintaining a natural balance in our environment. On February 10, 2017, the U.S. Fish and Wildlife Service listed the rusty-patched bumblebee as an endangered species, now found in only 13 states. Bumblebees and other pollinating insects help pollinate 75 percent of the fruits, nuts, and vegetables we eat and removing them from the equation could trigger a global food crisis according to the United States Department of Agriculture (USDA). The short term solution could lie with drones.

That's the size of the workforce dedicated to a business that's thrived for the past three decades on a growth model of labour arbitrage. In the process, it's matured into a industry worth almost $150 billion. The multi-billion dollar question today is, where does it go from here, as it comes head to head with a series of challenges. Consider: New Bills being introduced by the Trump administration in the US Congress seek to make outsourcing not only tougher but more expensive. This comes at a time when protectionist sentiments that call for onsite hiring of locals are spreading globally. All this, even as traditional IT services are moving towards the digital economy.

In this week's This Feels A Little Like Skynet: IBM built a new voice assistant using artificial intelligence called Hayvn, focused on cybersecurity. Think of it as Amazon Alexa, but instead of ordering soap, it's helping you manage threats. Sure, this might sound like it's ripped straight out of the plot for Terminator 3: Rise of the Machines, in which the military unleashes a new AI called Skynet to fight a virus that's been disrupting worldwide networks. And yes, that AI then becomes sentient, launches nukes and begins "Judgement Day." Here in the real world, IBM said Hayvn is being used to help cybersecurity pros comb through the hundreds of alerts they receive each day.

Two senators said Monday that they're launching a bipartisan effort to help to speed up the deployment of self-driving cars on the nation's roads. It's the first major congressional attempt to address the advent of the vehicles. Sens. John Thune (R-S.D.), the chairman of the Senate Commerce, Science and Transportation Committee, and Gary Peters (D-Mich.) said they're exploring legislation that "clears hurdles and advances innovation in self-driving vehicle technology." The senators' counterparts in the House are also gearing up to address the new technology, with a hearing scheduled for Tuesday. Automakers cite federal requirements that all vehicles must have steering wheels and brake pedals as examples of regulations that presume there will be a human driver and might inhibit the introduction of self-driving cars.

How humanity will meet its end is a an endless source of fascination in science fiction. But scientists claim many of the scenarios depicted in films - such as an asteroid strike and killer robots - may not be as far fetched as you might think. Now researchers at Cambridge University's Study of Existential Risk (CESR) have come up with a list of 10 threats that may some day trigger an apocalypse. Humanity faces an uncertain future as technology learns to think for itself and adapt to its environment. Artificial Intelligence, disguised as helpful digital assistants and self-driving vehicles, is gaining a foothold and it could one day spell the end for mankind if allowed to develop without strict controls.

Yuval Noah Harari might be Silicon Valley's favorite historian. His last book, Sapiens: A Brief History of Humankind, which detailed the entirety of human history and how Homo Sapiens came to dominate the Earth, was blurbed by President Barack Obama and Bill Gates, and Mark Zuckerberg recommended it for his book club. And more than 100,000 students have taken Harari's online course. In his new book, Homo Deus: A Brief History of Tomorrow, Harari looks forward and hazards a few guesses on what comes next for humanity. These next chapters in our history range from the utopian to the horrific, he says.