AI-based chatbots might have a larger scale and faster influence on the world than nobody expected.

 AI-based chatbots might have a larger scale and faster influence on the world than nobody expected. 

Chat GPT and its competitors have already changed the programming industry. Even their freeware versions can create many things faster than human programmers. Even if the AI-based chatbot cannot dump data straight into a file, it can show the example code. Then the human user can copy-paste that code to the programming editor. Of course, there is some kind of changes like database names and file paths, that the programmer must change. 

But also freeware versions of those AI chatbots can boost the effectiveness of the programming. When people ask which of those chatbots is the best, I must say: decide yourself. Even social media applications have some kind of AI-based chatbot extensions. The AI can create complicated programs very quickly by using public databases. But machine learning makes those systems even more powerful. 

The code above is created by Bing (free version) using the command: "Show me a database connection in C++". The complete code is below this text. And you can see how small changes there must be made. The C++ examples include the same code, but Bing finds it faster than human users. That makes it an effective tool that can improve the effectiveness of programming. 

A learning machine is like a worker who keeps the tools that the person often requires closer than tools that are not needed so often. 

In programming machine learning means that the system will store the modules and code that it needs in the shortcut. The shortcut can be the storage on a hard disk. Or it can be a database with a list of home pages that the system requires very often. There is storage about the user interface models, database connections, and SQL databases that the system normally uses. And that makes it faster. 

When a self-operating AI chatbot makes a program it can change the names of databases and their connections in each program. But the fact is that database connections are always similar. if they are written with the same programming languages. The only thing that makes a difference with those things in different programs is the names of databases. 

If the system keeps a book about codes and libraries, that it used in a certain type of software it could be faster and more effective. The AI-based chatbot works in the cloud. That means the users are using the same software that runs in the same entirety. That means the AI can download libraries and the code that it uses in some software into a "standby" position. 

If the AI must not use the public Internet in all its missions it would be faster and more accurate than any human can be. When we talk about "limited AI! we might mean the programming tool that can create the program by using locally downloaded libraries. In that case, the system uses the internet only for asking code models. Then that system can connect those models with the libraries that it needs. The thing that prevents the AI from making the software is that it doesn't have the required programming libraries.  

The cumulation of information makes the AI more versatile. Sometimes people ask: Can AI-based- chatbots reach awareness? There is the possibility that an AI-based chatbot can hire it's awareness from the users. But the thing that the AI-based chatbot can do is it can create new software at a fast speed. And that thing means the AI-based chatbots are the next-generation programming tools. 

https://techxplore.com/news/2023-09-chatbot-self-awareness.html

*********************************************************************

// Standard C++ includes

#include <iostream>

#include <cstdlib>

// Include the Connector/C++ headers

#include "cppconn/driver.h"

#include "cppconn/exception.h"

#include "cppconn/resultset.h"

#include "cppconn/statement.h"

// Link to the Connector/C++ library

#pragma comment (lib, "mysqlcppconn.lib")

// Specify our connection target and credentials

const std::string server = "localhost";

const std::string username = "root";

const std::string password = "";

int main()

{

    sql::Driver *driver; // Create a pointer to a MySQL driver object

    sql::Connection *dbConn; // Create a pointer to a database connection object

    sql::Statement *stmt; // Create a pointer to a Statement object to hold our SQL commands

    sql::ResultSet *res; // Create a pointer to a ResultSet object to hold the results of any queries we run

    try

    {

        driver = get_driver_instance(); // Get a driver to use to connect to our DBMS

        dbConn = driver->connect(server, username, password); // Connect to the MySQL server

        dbConn->setSchema("test"); // Select the database "test"

        stmt = dbConn->createStatement(); // Create a statement object

        res = stmt->executeQuery("SELECT 'Hello World!' AS _message"); // Execute a simple query

        while (res->next()) // Loop through the result set

        {

            std::cout << res->getString("_message") << std::endl; // Print the result

        }

        delete res; // Delete the result set object

        delete stmt; // Delete the statement object

        delete dbConn; // Delete the connection object

    }

    catch (sql::SQLException &e) // Catch any SQL errors

    {

        std::cerr << e.what() << std::endl; // Print the error message

    }

    return 0;

}

Machine learning needs stimulus for making solutions.

 Machine learning needs stimulus for making solutions. 

"A barren plateau is a trainability problem that occurs in machine learning optimization algorithms when the problem-solving space turns flat as the algorithm is run. Researchers at Los Alamos National Laboratory have developed theorems to prove that any given algorithm will avoid a barren plateau as it scales up to run on a quantum computer." (https://www.lanl.gov/discover/news-release-archive/2021/March/0319-barren-plateaus.php)

Have you heard of a "barren plateaus" problem? The name of that problem is coming from the J.R.R Tolkien book. Wherein the fictive Middle land is the very dry and hot place. The image, above this text, introduces the "barren plateaus" problem very well. 

In the next example, the food and water are the information. If a creature lives in the "fresh plateaus" it can take nutrients from nature. There is a bigger chance to make mistakes. But the nutrient is versatile and finding and testing new things makes the creature in the work. When a creature searches for things from nature there are many possibilities to test which type of vegetables or other nutrient sources the creature uses. Of course, that thing requires sometimes rise to the mountains. 

In that image, the problem is the mountain. The thing is that if the creature lives on the landscape at the higher image. That creature has stimulus. The green landscape offers motivation and the grass is the food and the creature wants to go to the mountain. The grass and water are everywhere and the creature has the motivation to rise to the mountain. And that could be willing to see the farter places. Or maybe the creature wants to get fresh air. 

The lower image introduces the situation. Where the creature lives in the "barren plateau". The water and food are in a pocket or bag and of course, the creature never makes mistakes if that creature wants to get a certain sandwich. The creature knows which pocket that creature can get sausage sandwich and where is the drinking bottle. But sooner or later, the nutrient would turn unilateral. In barren plateaus problem, the creature will get pre-made food. 

So if we are transferring that model to the information technology the creature will get pre-made solutions that fit in certain situations. And that makes this kind of model very limited. The situation is like that creature lives in the desert or "barren plateaus". The supporter brings water and sandwich to a certain point at a certain time. The food is guaranteed but it's always the same. 

And what the creature gets depends on the supporter. If the supporter wants to give the sausage sandwich that is the food. If someday the supporter wants to give the cheese sandwich that creature will get a cheese sandwich. 

When everything is pre-made the creature doesn't want to try itself to find food. There is difficult to make mistakes if some other person makes the food. The same way is in data science. If all problems are pre-solved that thing means that it's very hard to make wrong solutions. 

The term "flatten landscape" means that when the creature is living in "barren plateaus" the limited information sources makes the problems look harder to solve. Because the creature always is at a certain point where the supporter will bring a sandwich and water the creature is not even trying to climb mountains or solve the problem. 

"A barren plateau is a trainability problem that occurs in machine learning optimization algorithms when the problem-solving space turns flat as the algorithm is running".

"In that situation, the algorithm can’t find the downward slope in what appears to be a featureless landscape and there’s no clear path to the energy minimum. Lacking landscape features, machine learning can’t train itself to find the solution". (LosAlamos National laboratories, Solving ‘barren plateaus’ is the key to quantum machine learning)

https://www.lanl.gov/discover/news-release-archive/2021/March/0319-barren-plateaus.php

Image:https://www.lanl.gov/discover/news-release-archive/2021/March/0319-barren-plateaus.php

Deep neural networks are revolutionizing astronomy and many other things.

  

 Deep neural networks are revolutionizing astronomy and many other things.

Deep learning means that the person would know how to act in certain situations. "Learning" means. The person would know how to in any situation that connects with a certain case.

But deep learning means that the actor also knows why something is done in a certain case?  And also, the actor can predict some situations. In the computer world, the prediction means that the solution like some movement series will upload to RAM (Read Access Memory for) immediate use. 

Deep neural networks in the service of astronomers are opening the road for a new type of artificial intelligence. The idea of the deep neural network is searching the exoplanets is that the system is following a certain well-known exoplanet using multiple different types of telescopes.

And then, that system would make the database about the observations. Then the network is trying to look for the phenomenon that is matching with recorded data to other stars. The ability to collect the data also by using the smaller telescope simulates the targets. Those are farther in the universe than the well-known exoplanet. 

So the small telescope can simulate the situation where some exoplanet locates very far away from the Earth. Then this data can use as the data matrix for the larger size telescopes. And by using the larger telescope. The AI can see if the data that is collected from other and more distant stars match the data. That collected from closer stars. 

The model where the system makes the matrix about the case and then transfers that to other cases would be revolutionary in astronomy. The fact is that this kind of solution can benefit also in many other situations like AI-controlled cars or similar things. If some solutions would be a useful thing. The computer is loading the data from the environment and then the computer would load this matrix in its memory. 

And then in the future, when the automatic would drive to a similar environment the system would load the case matrix to the RAM (Read Access Memory) that the system would use that data matrix or solution immediately. That means the image of the environment or the position on the map would act as a trigger. That uploads the solution for immediate use. This is the computer-world version of the learning. 

So the learning system can predict the situation. When the system records the place, And if there is a match with some case that system preloads the solution or the movement series to its RAM. And that thing means the machine learns to predict things. Machine learning is the thing that is one of the most interesting things in computing. 

In a chess program, machine learning means that the system can record the games of the masters. And then it would make the database for each movement. Then the system must just make the counter-action against the opponent simply by reconnecting the databases. That thing makes the machine can learn things. Like what buttons are the most effective against a certain player. The AI can calculate how often some chess player is moving a certain button. 

And conclude what kind of role the button is the button for that certain player. The chess game is a useful thing to test how to connect databases. 

But the same way the ability to make spontaneous connections can test by using astronomical objects as the base. And that kind of system can help to benefit telescopes, chess, and the AI for finding the new and more independently operating robots and AI a reality.

https://scitechdaily.com/a-whopping-301-newly-confirmed-exoplanets-discovered-with-new-deep-neural-network-exominer/ 

https://thoughtsaboutsuperpositions.blogspot.com/

Machine learning brings social skills to robots.

 Machine learning brings social skills to robots. 

Things like social skills are important for robots that are working with humans. The skills like saying "pardon me, I'm hurrying" are interactive methods to tell that robot has some kind of mission. If we are thinking the case that robot works in warehouses and collect the stuff from the shelves. 

There is needed to tell people if they are closing the road. As an example the case that the robot collects food. That requires cold storage robot should tell people that it is in a hurry or its load is warming. So the people should step away from it, or the food is going bad. 

That thing means the robot has not had the entire day. To wait for those people will step away from its route. This is one example where the robot should communicate with people. In some other examples, the people are standing in the working area of the welding robots without protective masks. And that means robots should tell that they should use dark glasses because of the welding radiation. 

But when we are thinking possibility that artificial intelligence supports the people in the cases like studies. There is the possibility that it would help the student to remember things like how to mark sources. And what kind of standard the student must use in the work. But the artificial intelligence can also follow what kind of sources the student will use for the thesis. 

That thing means that the AI recognizes the type of the sources and that they will fit to work. Then the AI can discuss with the student what are students' opinions of the sources are. The AI can also search how long the source has been open, and if there is something that should not be, the AI can tell that thing to a student. 

The use of machine learning is free. There are no limits in areas where that kind of system can use. The things like networked systems are also offering those systems to share their experiences. Things like choosing the route for driving would be the thing where the AI can give a piece of advice. The GPS and non-stop updating real-time data from the traffic control would make driving more comfortable. And if the system knows where is planned to stop. 

And other kinds of things that thing can also make driving safer and more economical than they are today. The accuracy and mass of data in the computer are determining the accuracy and how effective artificial intelligence can be. 

This kind of system bases the idea that the computers of the vehicles can exchange information about the times, what they are using for some trip. That thing means that the cars can change information and the traffic control computers can help to select the routes that are the most economic and fastest. The thing is that if the system knows the destiny of every vehicle and the driving speed on roads. 

The system can turn part of the traffic to other roads avoiding the rush. There are, of course, people who need to use a certain road but if the computer can route traffic with different routes that thing can make the driving more economic. Artificial intelligence can also give warnings if the tank or battery of the vehicle is empty. 

The system can search the next recharge point but if the destiny is near the system can also tell how long it would take to load the vehicle at the destiny. For making this kind of process the most comfortable as possible. The system requires so much data as possible. Things like stops and fast visits. Should load to the computer. And also the possibility to reload the battery in those places. Must be stored in the computer's memory. 

https://scitechdaily.com/new-machine-learning-system-gives-robots-social-skills/