At the speech stage of the CCF-GAIR conference, I'm afraid that Michael Wooldridge, the head of the Department of Computing at Oxford University, was the only guest who sat in a speech. He sat down on the steps of the stage one third of the time.
Michael Wooldridge, Head of Computing, Oxford University
When Lei Fengwang (searching "Lei Feng Net" public concern) asked why he sat on the floor, he just shrugged and showed an iconic smile. "This is just a subconscious action. I feel it will be more comfortable."
But for whatever reason, Wooldridge's little trick virtually brought him closer to the audience. This "equality" move also implies a metaphor of artificial intelligence: although AI has been able to accomplish human excellence very well. It's a task, but it still hasn't been able to interact and talk on an equal footing with your friends, whether it's understanding a philosophical story, or getting a joke from you.
Of course, we cannot demand too much of artificial intelligence - although it has been developing for 60 years, it is still a baby in the long history.
The conscious mind is not that importantThis year's hottest artificial intelligence is DeepMind's Alpha Go, and Wooldridge also explained Alpha Go as a representative to explain neural network based artificial intelligence in the CCF-GAIR speech.
Artificial intelligence based on neural networks can be said to be the first to handle complex problems, but its shortcomings are also evident. Just as our neurons have no way of telling us how it itself has been calculating the results of decisions, artificial intelligence has no way to communicate its own decision-making ideas. Another research method of artificial intelligence is based on the logic of language. Symbol-based artificial intelligence can achieve "upgrading" of machines at the cognitive level, but as Wooldridge said, artificial intelligence lacks knowledge and lacks empathy as humans, which also makes strong artificial intelligence a long way forward. .
Wooldridge told Lei Fengwang that he believes that artificial intelligence with consciousness in the short term is not yet available.
“Consciousness is a very personal thing. Even if we can see the structure of the living brain, we still can't see how people think. This is also the difficulty of human beings unable to copy consciousness into artificial intelligence.†Although humans can Using MRI to scan the brain for activity, we are still ignorant about the composition of consciousness. "This is indeed a long road."
The learning of neural networks helps us understand the thinking of human decision-making, so on some weak AI, tasks with specific goals will make artificial intelligence more applicable to human life. "I don't think there is a need to copy a brain," Wooldridge said. "So our research, now is not more about letting machines learn to think like humans, but letting machine agents or software agents understand our natural language. Emotions, which are perhaps more important and practical for our communication with the machine."
From Spielberg's "Artificial Intelligence" to the "Mechanic Ji" in recent years, many film and television works will use humanoid robots as a manifestation of strong AI, and humanoid robots have also become the goal of unremitting efforts of some scientists. Sophia from Hanson, Nadine, Geminoid chat robot, and Robot Actroid imitating human are examples of humanoid robots.
And for whether it is necessary to make the robot look like a human, Wooldridge said he doesn't catch cold on humanoid robots. It is Wooldridge's concern that whether it can truly be accepted by people more than the appearance of humans.
"If we can work collaboratively with robots on the same platform and it can communicate with humans smoothly, then whether it has a human form is actually not that important."
AI allows us to rethink social relationshipsFor the continuous progress of artificial intelligence, the threat theory represented by Hawking and Musk is no longer a new topic. But before worrying about whether AI will exterminate humans, Wooldridge considers more about how humans are after the emergence of artificial intelligence. Deal with the relationship between self and society.
Currently, there are 3.5 million truck drivers and 500,000 taxi drivers in the United States. According to Wooldridge's prediction, these jobs will gradually be replaced in 20 years, but just as the automation wave in the 1970s has caused many people to lose their jobs, today’s labor Intelligence also brings the same problems, but he believes that the current AI is more likely to be applied in place of some specialized skills.
“A typical example is the legal secretary. The main content of this kind of work is to find the corresponding laws. Although this work has a high degree of familiarity with legal knowledge, it can also bring decent income, but it can be done using artificial intelligence. Better. I think in the future, maybe their work will be threatened by artificial intelligence."
Therefore, this has also brought a problem to the government: how to be responsible for those types of work that are eliminated in the wave of technology. In the process of gradual replacement, the contradiction between artificial intelligence and humanity is also a great challenge. In this regard, Wooldridge did not give a clear answer, but he proposed that the issue of job matching could be improved from the perspective of increasing education and technology investment. Artificial intelligence also allows us to rethink human social relations.
Artificial Intelligence Next: MedicalAt CCF-GAIR, Wooldridge and Professor Yang Qiang from the Hong Kong University of Science and Technology made a wonderful conversation. During this time, Wooldridge thought that medical and health would be the next direction of artificial intelligence.
In the field of diagnostics, the AI ​​community has had a lot of gratifying results: DeepMind has already reached a cooperation with NHS and will develop a machine learning system for identifying visual diseases together with Moorfields Eye Hospital; Watson has also successfully diagnosed one patient for the first time in recent days. Rare leukemia.
In addition to diagnostics, Wooldridge believes that long-term monitoring and scientific guidance are important directions for AI+ healthcare.
When we have a series of wearable devices such as Apple phones and fitbit, they can monitor human body indicators such as heart rate, steps and blood glucose levels in real time, and if these data can be used by AI, they can help us to Physical conditions to make judgments, such as whether you need to rest, exercise is enough and so on. "I believe these will make us understand our body better and become healthier, just like having a private doctor."
Wooldridge is good at the multi-agent field but has not mentioned it in his speeches. As for how multi-agent can be combined with the sports and health field, Wooldridge gave Lei Feng a simple example:
When the user monitors his/her physical condition at home, the data stored in the mobile phone can be synchronized to the doctor in real time. When there is a problem with the body index, the system can automatically coordinate the time for the user and the doctor and provide a function for making an appointment for the visit.
However, most of the current wearable devices have not yet been certified by the FDA and other related qualifications. Before that, the data they obtained cannot be used as the basis for medical diagnosis. At this point, Wooldridge believes that the current combination of wearable devices and artificial intelligence is mainly to provide a personal assistant, help users set daily goals, monitor the completion of the situation and give feedback, which requires artificial intelligence to be carried out on the one hand Objective monitoring, on the other hand, should be able to constantly adjust according to the actual schedule and physical conditions.
At the CCF-GAIR conference, Wooldridge also mentioned that the privacy of data needs to be treated very carefully. “At present we have not found a suitable way to maintain the balance between data privacy and public information, but this is an unavoidable and very Important issues, we are also facing challenges."
Wooldridge mentioned that both the United States and the United Kingdom have an increasingly serious problem of aging. Although artificial intelligence assistants are not comparable to human care, they are better than nothing. "This can be regarded as a kind of social progress," and at the same time, Wooldridge also mentioned that electronic devices such as smartphones are beneficial to the elderly to communicate and communicate without leaving home.
"Now some people worry that artificial intelligence assistants will be harmful to human normal communication, but I don't think this kind of concern is necessary." Wooldridge thinks that the more important aspect of artificial intelligence is that it has changed the way people interact and has strengthened human behavior. The relationship between individuals living and others.
Conclusion: Although Wooldridge's journey has been very full, but he also patiently answered every question of the editors - just like a friend. Although artificial intelligence cannot become our true "friend" in a short time, the scholars who focus on AI like Wooldridge will always be pioneers in our exploration of science.
A new rule from the Drug Enforcement Administration (DEA) threatens to upend the American hemp industry, and could even result in criminal prosecutions for manufacturers of CBD and delta-8 THC products.
The DEA says the [interim final rule," issued Aug. 20, is simply a matter of adjusting its own regulations to account for changes to the Controlled Substances Act that were mandated by the 2018 Farm Bill (or Agricultural Improvement Act) that legalized hemp and CBD production. The new rule [merely conforms DEA`s regulations to the statutory amendments to the CSA that have already taken effect," says the agency. The new rule doesn`t break any ground, according to the DEA.
But many experts on cannabis and hemp law say the DEA rule creates a potential pathway the law enforcement agency could use to prosecute hemp processors and producers of CBD (cannabidiol) and delta-8 THC (or Δ8THC) products. There are two issues: partially processed CBD, and [synthetically derived" delta-8 THC.
Cbd Pod System Oem,Cbd Vape Pod Oem,Best Cbd Pod System,Cbd Pod System
Shenzhen MASON VAP Technology Co., Ltd. , https://www.masonvap.com