Google announced earlierOpened its machine learning database framework "TensorFlow", allowing more developers to freely connect huge machine learning resources and apply them to various application services. Behind the development of this strategy, Google naturally hopes that by having more people join in using this resource, the computer system can learn increasingly complex usage behaviors.
Technology companies, including Google, have been investing in the development of machine learning for quite some time. From the simple comparison of input information and results in the past, to the subsequent addition of more variables, correlations and other items, the entire machine learning system must spend more time on "training". Therefore, the subsequent development began to develop the introduction of hardware resources such as GPUs for GPGPU parallel computing, so that complex data analysis can be completed in a shorter time.
However, since machine learning systems themselves still require extensive data analysis and comparison to achieve "correct" learning results, they inevitably require extensive training. Furthermore, different comparison data is added to accommodate different service categories and usage scenarios. Training solely through Google itself still has limitations in terms of accuracy. Further opening up the machine learning database framework through open source for more developers to leverage, not only would it drive the development of more services that utilize machine learning, but it would also allow the database to become even smarter by accumulating training experience across diverse scenarios.
As for Google's current development strategy, it is still mainly to make the computer system more convenient to use.Currently notThe goal is to create an artificial intelligence system that can think for itself and understand user emotions, but this does not mean that Google has not considered such a goal. Therefore, it has chosen to open the database for free use by everyone. After being able to learn logical judgments in various usage scenarios more quickly, it is expected that it will be able to learn various skills by memorizing a large amount of knowledge, just like the human brain.
With the current relatively forward-looking artificial intelligence technology, the actual thinking and learning level is about the same as that of a 4-year-old child. Overall, it is certainly not "smart" and there is still a considerable gap from our imagination in real life. It is expected that Google's open access to its machine learning database framework "TensorFlow" will further drive the development and growth of artificial intelligence systems.
However, it's unclear whether opening up its machine learning database will benefit Google's technological development. After all, the "experience" accumulated over a long period of time in machine learning still needs to be combined with front-end computing systems to provide better services. In other words, if sufficient "experience" is possessed but it cannot be properly utilized, it will still not constitute more "smart" artificial intelligence. This part depends on the design skills and actual application models of various manufacturers and developers.
Whether we will see the emergence of artificial intelligence that surpasses the human brain and may even pose a threat to humans in the future may depend on the "Three Laws of Robotics" emphasized in science fiction novels and movies in the past.Note)" Add relevant ethical standards to artificial intelligence to prevent it from having an unauthorized impact.
Note:The Three Laws of Robotics are a set of behavioral guidelines for robots, outlined in science fiction novelist Isaac Asimov's 1950 novel "I, Robot." These guidelines include the first principle, "A robot may not injure a human being or, through inaction (or by inaction), allow a human being to come to harm," the second principle, "A robot must obey human orders except where such orders conflict with the First Law," and the third principle, "A robot must protect its own actions unless such actions conflict with the First or Second Laws."
