무위와 불교의 중도, 그리고 인공지능의 알고리즘에 관한 고찰

A study on dynamic coherence among the Tao Te Ching, the Middle Path and the AI algorithm

본고는 고대 동아시아 사유 방식의 한 축을 담당하는 『도덕경』의 무위사상과 『중론』의 삼제게 및 현대의 인공지능 알고리즘의 베이즈 정리 사이에는 상호 공유되는 상즉의 사유방식이 존재할 것임을 전제하고 이를 고찰하는 것을 목적으로 한다. 이는 여시아문의 ‘문(聞)의 관점’을 여시아전(如是我轉)의 ‘전(轉)의 관점’으로 새롭게 재구성하여, 여시아현(如是我顯)의 ‘현(顯)의 관점’으로 중도의 내재적 가치를 드러내는 일이라 하겠다. 이를 위해 첫째는 『도덕경』의 제1장에 근거하여 도는 가(可), 즉 그 특성이 가변적임을 명확히 함으로써, 도가 무유상즉의 상호의존관계로 작동되고 있음을 밝힌다.무는 체용의 체로서 삼제게의 공제의 특징을 지니고 있으며, 유는 체용의 용으로서 가제의 특징을 지니는 것으로 배대할 수 있는 가능성을 제시한다. 또한 무위무불위의 무위는 공제로, 무불위는 가제로 비정할 수 있어, 도는 중도의 원리와 상응하고 있음을 추론할 수 있다. 둘째는 의식지향체계로 규정할 수 있는 베이즈 정리는 주관을 버리고 객관으로 끊임없이 사고의 틀을 개선시키는 학습방식이라는 점에서 중도적 사유체계와 상응하고 있음을 알 수 있다. 베이즈 정리의 사전확률은 분별적 사고 틀로 해석할 수 있으며, 전확률은 공제로, 우도확률은 가제로, 그리고 사후확률은 중제로 비정할 수 있다는 것이 본고의 추론이다. 인공지능의 딥러닝 및 강화학습방식이 중도 사유체계와 상호 맞닿아 있음을 확인할 수 있다.

The purpose of this paper is to consider the premise that there will be a strategic framework for a shared way of thinking among the Muwi(無爲) thoughts of the Tao Te Ching, which play a role in the ancient East Asian way of thinking, and the three fundamental truths of the Middle Path and the Bayes Theorem of the artificial intelligence algorithm. It is analyzed and suggested that the Buddhist system of thinking in the Samje Weonyung(三諦圓融) is in close contact with the ideology of the Tao Te Ching and the Bayes Theorem, one of the learning algorithms of artificial intelligence. This is an attempt to explore the convergent implications of the Buddhist way of thinking inherent in the scriptures represented by Yeosiamun(如是我聞), that is to say, from the current perspective, it is an attempt to reconstruct the view of the Hearing(聞) of Yeosiamun(如是我聞) into the wheel of the dharmacakra-pravartana(如是我轉) in line with the innovation of modern science and technology, as well as to reveal the strategic value of the Middle Path with the manifestation of Yeosiajeon(如是我顯). Based on those recognition, this study newly present the strategic value of the Buddhist thinking system from two perspectives. First, by revealing that in light of Chapter 1 of the Tao Te Ching, the Tao(道) is referred to as having the characteristics of Ga(可), and it is clarified that Tao's properties are variable, and that Tao operates as a significant interdependence between Mu(無) as a potential world and Yu(有) as a phenomenal world. In this way, it is judged that the principle of operation of Tao(道), which works as an interdependence between Mu(無) and Yu(有), can be easily passed on as a thinking system of the Samje Weonyung(三諦圓融). Tao(道) is defined as a Ga(可), that is, change itself, which is equivalent to the principle of Muwi-MuBulwi(無爲無不爲). It is reasonable to estimate that it also can be interpreted as the three principles of the Middle Path that play a major rule as a strategic framework for the Buddhist thinking system. Second, Bays Theorem, one of the AI algorithms defined as another consciousness, is also a decision-making theory that abandons subjectivity and moves toward objectivity, not just a computer operating system. It has been shown that the process can be subsumed into the three principles of the Middle Path in that it is a learning method that constantly modifies the intellectual system of consciousness based on observation, data, and events. The reasoning of this paper is that the prior probability of Bays Theorem can be represented as a framework for divided thinking, and that the total probability can be determined as a Gong(空), the likelihood probability as a Ga(假), and the posterior probability as a Jung(中). It can be said that it is a process of dismantling the discriminated cognition system caused by the fixed frame and fringe limitations through Gong(空), presenting an updated direction based on this, and realizing the Middle Path. It can be confirmed that the same mechanism of operation as AI's deep learning and reinforcement learning methods interact with the Middle Path thinking system.