Explore how humans learn their own language，It has always been a very attractive research topic in linguistics and neighborhoods，The name of "Plato" by researchers at home and abroad。Psychology theory of behavioralism in the 20th century，The theory of cognitive development to Piajie、Jamesky's language internal theory，and the current rising machine learning exploration，There are different opinions on answering answers to how humans learn the question of language。This difference makes linguistics、Psychology、Related research in the fields of language philosophy and other fields embarked on different development paths，It also prompts us to reflect on what is the revelation and lesson of the current rapidly promoting machine learning research.。

　　Bet365 lotto review

Generally speaking，According to different points in the learning process，How to process input data and algorithms，Machine learning has different categories。If the learning method is closely related to language，Can be divided into supervision learning、Unsupervised learning and strengthening learning。Among the three，Reinforcement learning is closely related to the behaviorism of the field of psychology and language acquisition in the mid -20th century，Follow more attention。

One of the founders of enhanced learning Richard Saton believes，Reinforcement Learning is how to map the environment and behavior，To maximize the reward signal (enhanced signal)。Based on the expectations of possible rewards，Decisions that affect behavior。During the interaction with the Bet365 lotto review environment，Learners do not know what action should be taken，What kind of behavior must be summarized by continuous attempts to get the maximum reward。So，Trial and error search is one of the significant features of strengthening learning。Because of the useful information, you can get the biggest reward expectation，Make favorable decisions。，Under a specific situation，Some behaviors not only affect direct rewards，also affects the next link and all subsequent rewards generated by it，Delayed rewards have become another prominent feature of strengthening learning。Visible，Just as an expert in artificial intelligence research expert Dalis Kazakhstis recently explained in the "Nature" magazine，The entire interaction process includes behavior、Reward expectations and decision -making。

　　Human learning language is a interpretation process

Considuous language learning theory dominated by behavioralism，Language learning is the result of a learner's response to external stimuli。Based on this，Language environment and stimulus strengthening the decisive role of language learning。This is because of，Human brain can form memory to stimulate external language，so as to master the language。As described above，The basic concept of strengthening learning is to use reward expectations (similar stimulation) to strengthen the correct behavior。，When the reward prediction error is approaching zero，Reinforcement learning reaches the best state。In this sense，Reinforcement learning can also be said to be trial and error，The prediction error of minimizing the prediction。What is more noticeable is，Kazakhsbisbis and others recently tried to explain the learning algorithm in "Nature" magazine to explain the learning process of the human brain，Distributed bet365 live casino games enhanced learning under artificial intelligence perspective，It seems that you can rely on dopamine and other neurotransmitters that can promote the realization of rewarding expectations，Implementing it in the human brain。In short，If the behavior meets the reward expectation，Get positive dopamine signals，Then promote the correct learning decision。

But，Human learning language is just like the theoretical claims of the Jimsky language learning theory，It is a interpretation process。Learners propose some kind of hypothesis when learning the language structure，Then the assumption test is performed according to the original language data input，Correct or confirmed hypothesis about the target language structure，Complete language acquisition。This process seems to be similar to the trial error summary learning in learning，But it is the verification process of previous hypothesis，instead of inductive learning process。Especially for children's mother tongue，Alclaration summary learning has been questioned，Explain language acquisition from the perspective of biological maturity from the internal language mechanism，It has been confirmed by more neurological and experience research。

But this does not specifically explain the specific process of children's quickly obtained mother tongue。In other words，Language itself as a very complicated self -sufficient system，Learning theory of this complex system，It must be explained seriously how children are fast、Consolidated Language Body Properties such as structural hierarchy。Although dopamine such neurotransmitters are indeed in the process of language acquisition, the relevant nerve implementation，But this is just a specific manifestation of the neurological biological foundation Bet365 app download obtained by language。More important question is，How to get the specific process of language acquisition in the neurological and biological foundation based on language。

again，Although strengthening learning, there is a process of examining language learning from a psychological perspective，3573_3600，With super high capacity and super strong computing capabilities，It is worth considering。From the perspective of the research on the neurological and biological attributes of the brain，Cognitive computing of human brain is far less than machine。So，The difference between human language learning and algorithm -based machine learning is difficult to eliminate。

　　Mutual learning and learning is feasible

Exploring language acquisition process involves the synergy of multidisciplinary。mutual promotion and mutual learning between disciplines is the reason for the rise of cognitive science。Strengthen the interpretation of language acquisition questions，Both computer science、The characteristics of the cross -disciplinary and other disciplines of neuroscience。Discovery，Then integrate language philosophy、Linguistics、Neuroscience、Biology and Psychology Expand。Visible，Both learning and mutual learning have natural possibilities。

Well knows，Reinforcement requires a lot of data training to achieve tasks，and children quickly、Unanimously obtaining language does not seem to need a lot of language training，On the contrary，Facing the facts that stimulate the poor。Even the language outside the mother tongue outside the adult，In addition to working hard to remember vocabulary，You can also master grammar without a lot of training。Language learning depends mainly on the inner language mechanism。This means，The doubts encountered in the initial stage of machine bet365 best casino games learning research seem to have not been effectively resolved: relying on a large number of algorithm training, learning may be essentially different from human learning language。Even under the rapid development of the current algorithm technology，Such as the development of scene memory and meta learning technology，This difference does not seem to be a substantial breakthrough。As some machine learning researchers ridiculed，Use a model to learn human language，Algorithms driving learning often only learn to be one repeat machine。In view of this，These two different studies are different in the rapid development of artificial intelligence technology，You can learn from each other，and jointly promote the discussion of the "Plato problem"。

First，Both theories support language learning is based on neurological biological foundations。Despite strengthening learning identity stimulus -reactive language learning concepts，But also accepts language learning to rely on brain neuroma and biological properties to achieve this consensus。Expectation of the ideal reward，Nervous cells release dopamine to promote the corresponding neurological activity。This process explains that learning does have a neurological basis，Therefore, the theoretical focus of the theory with the Language Language acquisition of the Language Language is not conspiring by the impact of the brain's neurological properties on language acquisition。Under the current minimally plan biological linguistics research paradigm，The core of the study of language habits becomes the basic language attribute determined by the brain's neurological biological attributes。

Although the internal basic language attributes are the basis for the expansion of language bet365 Play online games habits，But language learning needs to be carried out in the actual environment，Due to the construction and proficiency of the language system, the language processing operation is required to activate the basic language attribute、CICF is in the corresponding brain module。So，This process is related to the algorithm that strengthens the development of learning。Such algorithms similar to language learners rely on language processing operations to master the language structure，Familiar language skills。Visible，The main difference between strengthening learning and mainstream language acquisition theory is，Whether the preset language learning requires certain basic language attributes determined by biological gene attributes。You can learn from each other in language processing，Even exploring similar algorithms，As mentioned above, two very promising technologies that appear in the current artificial intelligence field -Scenario Learning and Yuan Learning，Although there is a difference between the human brain and the machine brain in the implementation of the specific algorithm。

In short，Just as Kazakhs and others believe that dopamine can motivate rewards and expect to promote learning decisions，Mainstream language acquisition theory has always believed that language acquisition or language itself has a solid neurological biological basis。and，Enhance learning experience in algorithm development，Help the current mainstream language acquisition theory to explore the in -depth language learner to master the language through familiar language computing。In this sense，Research and mainstream language habits research paradigm is a continuous unification，Feasibility of mutual learning。This shows to a certain extent bet365 best casino games the direction of the future efforts of machine learning and mainstream language acquisition theory。

　　(Author Unit: Academy of Foreign Languages ​​of Suzhou University)