For the first time, I've written an article about the terms "adaptive" and "non-adaptive" corruption that address the blockchain's vulnerabilities.
Then I found the same term on this ethereum github page (link to page: https://github.com/ethereum/wiki/wiki/Sharding-FAQ#what-are-the-security-models-that-we- are -operating-under) there is a very short definition of adaptable against not adaptive Opponent as follows:
an opponent is adaptable if you can fast choose which part of the validator is set to "corrupt" and not adaptive if only they can make that choice far ahead of the times,
But this definition is not clear to me. And only the Ethereum platform is considered.
For example, What Does that mean that an opponent chooses? fast Part of the validator should damage them? How to say, her choice was fast or not ?
I searched for this term in another platform and found many in several computer security articles. The main question is whether this "adaptive corruption" or this "adaptive adversary" is applicable to Bitcoin or not.
Note 1): To the best of my knowledge, the term "adaptive adversary" has been used in many scientific publications on computer security, but I have not yet been able to find a clear and precise definition for that term.
As an example, here's another quote from another paper like this: (There are many other similar examples.)
We note that some of the earlier works, such as Algorand and Ouroboros, are taken into account
a stronger one fully adaptive opponent you can choose that freely
controlled participants for each time slot. Our opinion is that such a
A fully adaptive opponent is interesting and worth the study, but often
not realistic. In practice, a platform compromise is difficult to detect and correct
Repair. In addition, a compromise of a computer platform is not possible
mean that someone is no longer in control of the opponent. For this
Reasons to focus on non-adaptive opponents in this work.
Note 2): I took up this question security.stackexchange and I received this answer:
"An adaptive opponent can adapt to the environment and the reactions of the environment as part of the attack – a non-adaptive opponent
that could mean that this term is a general term in safety. I have also found this term in many papers on computer security, and it is even used Non-block-based chain Fields. As a result, "adaptive opponent" seems to be one General Term and is not only used for the validator set,
Application at Bitcoin:
In my view, the last quote is a more precise definition base on which we can look at it for the bitcoin application. That is, a pre-written code (as a robot) for a given attack (such as egoistic mining) is a non-adaptive adversary, as it can not adapt to the new strategies of the honest miners in the area. But if this were an adversary flexible on environmental changes and new strategies, then we could consider it as one adaptive opponent,
So, according to the above statement in Bitcoin, can we say that an attacker who intends to damage the network can be either "adaptive" or "non-adaptive," based on his flexibility to adapt to environmental changes, such as: For example, the network, updates to be updated, changes in network hashing performance, changes in the strategy of honest miners, etc.?
I also found a definition in the following article that I think could be a general definition for any distributed and peer-to-peer network.
In this article we read:
Adaptive falsifications. The opponent influences the execution of the protocol to interact with the available functionalities, and from
corrupt parties, To corrupt a party, the opponent must first ask
the environment Z for a permit. If the corruption is approved by Z
(over a special message from Z to A), the opponent spoils Pi