Selfish mining attacks get a high prize due to the additional rewards unproportionate to their mining power (mining pools have particular advantages). Generally, this category of attacks stresses decreasing the threshold to maximize the rewards toward the view of attackers. Semi-selfish mining falls into the family of selfish mining attacks, where the threshold value is approximately 15%. However, it gets little attention to implement these attacks in practical. In this paper, we focus on the validity of semi-selfish mining attacks considering the probability of being detected. More specifically, we discuss mining strategies through backward deduction. That is to say that the attacking states derived from the observable states, which with normal forking rate, just as without semi-selfish mining attacks, toward the view of the honest miners. Rewards distribution is further investigated concerning these strategies. The simulation results indicate that it does not necessarily bring rewards advantage over large pools. Instead, the small pools have an advantage over the additional rewards. However, the probability for small pools to successfully implement these strategies is pretty low. That is, it is impossible for the pools, although profitable for them, to sponsor semi-selfish mining attacks without being detected.

中文翻译：

---

是否可以在不被发现的情况下进行半自私挖矿？

自私挖矿攻击因其额外奖励与其挖矿能力不成比例（矿池具有特殊优势）而获得高额奖励。通常，此类攻击强调降低阈值以最大化对攻击者观点的奖励。半自私挖矿属于自私挖矿攻击家族，阈值约为 15%。然而，在实际中实施这些攻击却很少受到关注。在本文中，我们关注半自私挖矿攻击的有效性，考虑到被检测到的概率。更具体地说，我们通过向后演绎来讨论挖矿策略。也就是说攻击状态来源于可观察的状态，在正常分叉率的情况下，就像没有半自私挖矿攻击一样，朝向诚实矿工的观点。对这些策略的奖励分配进行了进一步研究。模拟结果表明，它不一定比大型矿池带来奖励优势。相反，小型矿池比额外奖励更有优势。然而，小型矿池成功实施这些策略的可能性非常低。也就是说，矿池不可能在不被发现的情况下赞助半自私的挖矿攻击，尽管对它们有利可图。