To use chat, please login.
Back to contacts

Artificial General Intelligence V.6.0 (by leo23)

Прирост: +7.85M%
Просадка: 25.79%
Пункты: 18238.9
Сделки: 7407
Прибыльных:
Убыточных:
Тип: Демо
Плечо: 1:200
Торговля: Автоматизированный
bmigette
Dec 04 2018 at 07:29
63 сообщений
I'm sorry, but from your statements it looks like you're not really a machine learning developper.

A trained model size is irrelevant of the size of the training data, so I'm curious what do you mean by 'file sizes increase day by day.'. If you store history to re-train your model, obviously the history would grow up, but it would just affect the time it would take when you retrain your model, and should not affect current used model.

also, live training of your model is called reinforcement learniing, and I don't think you need to store any data for this kind of model....

Just sayin.

leo23
Dec 04 2018 at 07:54
552 сообщений
bmigette posted:
I'm sorry, but from your statements it looks like you're not really a machine learning developper.

A trained model size is irrelevant of the size of the training data, so I'm curious what do you mean by 'file sizes increase day by day.'. If you store history to re-train your model, obviously the history would grow up, but it would just affect the time it would take when you retrain your model, and should not affect current used model.

also, live training of your model is called reinforcement learniing, and I don't think you need to store any data for this kind of model....

Just sayin.

Machine learning is a vast topic to discuss in theory. We can keep on discussing the different types of machine learning like supervised learning, unsupervised learning etc and 'reinforcement learning' is just one type of machine learning based on agent reward which is somewhat like a feedback mechanism.

Yes, we are referring here a type of reinforcement learning along with deep neural network and so we can call it as 'Deep reinforcement learning'.

If you have already developed any machine learning algorithms and trained a model with past data of a couple of months or years and continue to retrain it everyday, then you might understand what we are referring to exactly.

So neither the model size and nor the past trained data remains constant after each training. But we try to disregard the old data as the time passes by, but we have to do it carefully as well to avoid curve fitting and that is what we are referring to as file size keep increasing means the size of the models keep increasing when we add more data for training on every new day.

Most importantly after a threshold point of model size, the system simply freezes and is unable to place any trades properly in time due to massive search space and hence, we have to either use GPU or switch to python to use TPU(Tensor processing unit) used by Google. Now, we are just using some API bridges from MT5 and third party libraries to connect to MT4.

By the way, we are not using any kind of learning in this demo account and this is just a simple neural network with inbuilt weights which we keep changing in the code from time to time. We are referring to machine learning when we use this algo in live accounts only.

If you have any ideas as how to tackle such problems in machine learning and use the algos with least computational power, then you are welcome to discuss here. We will be happy to share some of the most powerful ideas used in latest machine learning algos like Alpha zero used by google deep mind and which can be implemented in forex to develop a real AI which can almost never loose in forex. But everything ends up in freezing the mt4 or mt5 platform when we finally implement the algos and hence, without massive computational power it seems difficult to make it work perfectly in forex live trading though it may work sometimes in demo accounts.


trader (spring)
Dec 05 2018 at 08:07
17 сообщений
please show live account statement, and open history

leo23
Dec 05 2018 at 08:28
552 сообщений
spring posted:
please show live account statement, and open history

We may publish a live account for public access when we will feel we need to do so. We are already working privately in co-operation with other experienced traders. We don't have to make anything public to people who have no idea about our project.

We are not in a hurry to convince any client or attract any client to us and by doing so we don't want to just jump in to live account messing up our intermediate steps of testing. Also, we don't need any clients as well.

This is an experimental project based on 'Deep neural network' and 'Machine learning' which we need to thoroughly study in demo before putting it in live account, because probably this is the last project we will be working on and after that we don't have to do anything once the algo start learning and adopting by itself in the changing markets and can still work in a regular server.

Those traders who have knowledge in these fields are free to participate and contribute on this project and discuss in our thread. Our thread is not for EA buyers, investors or quick money maker newbies. People who have no knowledge on these should stay away from our thread and we don't want to waste our time with them.

Regarding trade history, we never shared trade history of any of our previous projects and same is true for this one as well. We publish only the trading performance and not the trade history.

leo23
Dec 06 2018 at 15:36
552 сообщений
For those of you who are using our EA we want to update that we have done some changes to the code which need to be updated. We have added some more parameters to be seen on the mt4 screen while the EA is running.

Most importantly we have optimised the weights of the neurons of individual hidden layers of our neural network. So the results might be a little improved and more adapted to the current markets though volatility will be low since we are in December now.So we might see less trading opportunity through out December.

The EA can be updated directly by restarting the mt4 terminal and it will automatically get updated within a couple of minutes in a regular server or VPS.

leo23
Dec 07 2018 at 17:42
552 сообщений
We have completed some of the initial phases of testing of our machine learning algo in demo accounts to trade during specific market conditions and we still need more time to improve the settings and coding to make it work fully automated 24 hours.

But we are almost ready to test it in a live account which we may do in coming weeks. However, as we mentioned in our previous post since we are in December and hence, naturally the liquidity in the market for many currency pairs will be less as well as markets may be flat most of the times.

Hence, it may not give an appropriate understanding about the true effectiveness of our algo. However, we will see if we can start a live account with small risk. We already started our live account previously and stopped immediately due to obvious reasons and we don't want to repeat the same again. So we are just delaying the live account testing for some more time.

Good Choice FX (goodchoicefx)
Dec 09 2018 at 08:15
2 сообщений
Hello,

Seems interesting, Which platform does this EA work on? MT4 or MT5?

Thank you

leo23
Dec 09 2018 at 08:23
552 сообщений
goodchoicefx posted:
Hello,

Seems interesting, Which platform does this EA work on? MT4 or MT5?

Thank you

This EA is only for MT4.

We are working to implement it to MT5, but only for the purpose of machine learning since currently it is difficult to implement and make machine learning work in MT4 properly and MT5 has better advantages in regards to machine learning.

But in case if we do that, then MT5 version will be only for our personal use and not for the public.

leo23
Dec 10 2018 at 16:51
552 сообщений
It has been a couple of months since we asked the users as how many of them will be interested to copy the trades from our system and so far we have got around 4 to 5 users requests for signal subscription and with long gaps and this is of no use for us.

We need at least 20 to 30 user requests for getting our signals and then, we will consider to provide signals from our system, because many of the users may or may not join or may leave after a couple of weeks for various reasons and we don't have time to waste for a couple of users and we are fine in providing our services privately for now.

So we would like to repeat that if anyone is interested in subscribing to the signals of our system, then make a comment in this thread showing your interest and if we will receive at least 20 to 30 user requests, then we may make this system available for signals.

PROFIT095
Dec 11 2018 at 07:51
7 сообщений
На каком сайте можно посмотреть ваш торговый робот ? И можно ли его купить ? Если да , то какая цена ?

Чтобы оставить комментарий, необходимо войти в систему.