【推荐1】Everyone becomes a little more forgetful as they get older, but men’s minds decline more than women’s, according to the results of a worldwide survey.

Certain differences seem to be inherent in male and female brains: Men are better at maintaining and dealing with mental images (useful in mathematical reasoning and spatial skills), while women tend to excel (擅长) at recalling information from their brain’s files (helpful with language skills and remembering the locations of objects).

Many studies have looked for a connection between sex and the amount of mental decline (衰退) people experience as they age, but the results have been mixed.

Some studies found more age-related decline in men than in women, while others saw the opposite or even no relationship at all between sex and mental decline. Those results could be improper because the studies involved older people, and women live longer than men: The men tested are the survivors, “so they’re the ones that may not have shown such cognitive decline,” said study team leader Elizabeth of the University of Warwick in England.

People surveyed completed four tasks that tested sex-related cognitive skills: matching an object to its rotated form, matching lines shown from the same angle, typing as many words in a particular category (范畴) as possible in the given time, e.g. “object usually colored gray”, and recalling the location of objects in a line drawing. The first two were tasks at which men usually excel; the latter are typically dominated by women.

Within each age group studied, men and women performed better in their separate categories on average. And though performance declined with age for both genders, women showed obviously less decline than men overall.

【推荐2】The atmospheric level of carbon dioxide (CO₂)—a gas that is great at trapping heat, contributing to climate change—is almost double what it was prior to the Industrial Revolution. This presents a challenge to researchers attempting to design artificial trees or other methods of capturing (捕获) carbon dioxide directly from the air. That challenge is one a Sandia National Laboratories-led team is trying to solve.

Led by Sandia chemical engineer Tuan Ho, the team has been working on a project using powerful computer models combined with laboratory experiments to study how a kind of clay can take in carbon dioxide and store it. “These fundamental findings have potential for direct-air capture; that is what we’re working toward,” said Ho. “Clay is really inexpensive and rich in nature, which should allow us to reduce the cost of direct-air carbon capture significantly, if this high-risk, high-reward project ultimately leads to a technology. What’s more, it can be used like sponges (海绵) to absorb carbon dioxide, and then the gas will be ‘squeezed’ out of and pumped deep underground.”

Carbon capture and storage is the process of capturing extra carbon dioxide from the Earth’s atmosphere and storing it deep underground with the aim of reducing the impacts of climate change such as more frequent severe storms, rising sea levels and increased droughts and wildfires. This carbon dioxide could be captured from fossil-fuel-burning power plants, or other industrial facilities or directly from the air, which is more technologically challenging. Carbon capture and storage is widely considered one of the least controversial technologies, to deal with climate change.

“We would like low-cost energy, without ruining the environment,” said Susan Rempe, a Sandia bioengineer on the project. “Direct-air carbon capture is important for reducing the amount of carbon dioxide in the air. We can live in a way that doesn’t produce as much carbon dioxide, but unfortunately, we can’t control what our neighbors do.”

【推荐2】Many cases of online bullying have been making the social media rounds in recent years. Negative moral judgment, cyber threats, and verbal violence(语言暴力) pollute the digital world. Meanwhile, some platforms, attracting network traffic, tend to turn a blind eye to the roaring of verbal violence and so the list continues to cause a great stir on social media from time to time.

There are 43.88 billion people surfing the Internet and spending a weekly average of 28.5 hours online. The question becomes: What measures can help build a good cyber environment and, more importantly, limit online violence?

While providing the public with more room to express and share their ideas, the Internet for some serves as a platform to hurt others. Online anonymity(匿名) makes it possible for some netizens to stop obeying all morals, as well as the related laws and rules. With their identities covered, netizens can be less worried to do what they want to do, thinking that they don’t have to bear legal responsibility. Cyber space is not lawless area. To create a healthy cyber space and limit online violence, real-name registration can prove effective. Law breaker and platforms that refuse to take responsibility should also be punished.

Good laws are what it takes to manage the present situation. To limit cyber bullying and clean the digital space, one key step is to strengthen lawmaking and law enforcement(执行). Prevention is better than cure. Strict law enforcement appears to be more important.

However, some countries still lack special laws targeting cyber violence, and only have a few terms and unsystematic rules in place-too weak to deal with the complicated violence online. At this point, though sometimes illegal acts like verbal violence will lead to a maximum of three years in prison, it can be costly and difficult for sufferers to collect evidence and safeguard their rights. The number of cyber bullying cases published is very low. There are a few technical difficulties here. It is hard to identify who should bear the blame and sufferers have no idea who they should file a complaint against.

【推荐3】First the robots came for our jobs. Now they’re coming for our hobbies. Google’s DeepMind AI lab has been busy the past few years, creating programs to take on human players across a variety of games. In 2016 its AlphaGo beat the best Go player in the world. Earlier this year, its Alpha Star defeated two middle-tier players at the popular online game StarCraft II. Now it’s learning to win at multiplayer games.

“Artificially intelligent agents are getting better and better at two-player games, but most real-world endeavors require teamwork,” DeepMind’s researchers wrote in a paper published in Science in June.

To be sure, computers have been proving their dominance (优越) over humans in one-on-one turn-based games such as chess ever since IBM’s Deep Blue beat Russian chess master Garry Kasparov in 1997.

However, successfully using teamwork to win in multiplayer games with complex environments was difficult to achieve.

For this purpose, DeepMind’s researchers designed AI agents that taught themselves how to play first-person game Quake III Arena. The team, led by Max Jaderberg, worked on a modified (修改的) version of Quake III Arena.

The game mode they chose was “Capture (抢) the Flag”. All the players must work together to steal the other team’s flag while safeguarding their own.

The AI agents “trained” with 12 hours of game data, then matched up against professional game testers. The AI won 75 percent of the time, even when its reaction time artificially slowed down to human levels and when their aiming ability was similarly reduced.

And in order to make improvements, the programmers used various kinds of “rewards” to help the AI players weigh their goals and actions to help them cooperate in the teamwork. They also used randomized maps for each new match.

“That meant the solutions that the agents find have to be general –– they cannot just memorize a sequence (系列) of actions,” said co-author Wojciech Czarnecki.

The only time humans were able to beat the AI agents was when they teamed up together. A team consisting of one human and one AI agent had a five percent greater win probability than a team of just AI agents.

Ethan Gach said at the video game news website Kotaku, “It suggests that the AI program is able to adapt to playing with non-AI teammates.” In the near future, AI may cooperate with human beings in some other fields, such as medicine and other branches of science.