| viXra.org > Artificial Intelligence > 2008 |
 |
 |
| viXra.org > Artificial Intelligence > 2008 |
|
|
[3] viXra:2008.0216 [pdf] submitted on 2020-08-30 09:55:52
Authors: Xiaozhuan Gao, Lipeng Pan, Yong Deng
Comments: 19 Pages.
Dempster-Shafer (D-S) evidence theory is an effective methodology to handle unknown and imprecise information, due it can assign the probability into power of set. Quantum of mass function (QM) is the extension of D-S evidence theory, which can combine quantum theory and D-S evidence theory and also extended D-S evidence theory to the unit circle in complex plane. It can be seen that QM has the more bigger uncertainty in the framework of the complex plane. Recently, negation is getting more and more attention due it can analyze information from the another point. Hence, the paper firstly proposed negation of QM by using the subtraction of vectors in the unit circle, which can degenerate into negation proposed by Yager in startand probability theory and negation proposed by Yin. et al in D-S evidence theory. the paper proposed quantum pythagorean fuzzy evidence theory (QPFET), which is the first work to consider QPFET from the point of negation.
Category: Artificial Intelligence
[2] viXra:2008.0163 [pdf] submitted on 2020-08-22 05:30:26
Authors: Shirui Tang
Comments: 12 Pages.
Preceptron model updating with back propagation has become the routine of deep learning. Continu-ous feed forward procedure is required in order for backward propagate to function properly. Doubt-ing the underlying physical interpretation on transformer based models such as GPT brought aboutby the routine explaination, a new method of training is proposed in order to keep self-consistencyof the physics. By treating the GPT model as a space-time diagram, and then trace the worldlinesof signals, identifing the possible paths of signals in order fot a self-attention event to occure. Witha slight modification, self-attention can be viewed as an ising model interaction, which enables thegoal to be designed as energy of system. Target is treated as an external magnetic field inducing sig-nals modeled as magnetic dipoles. A probability network is designed to pilot input signals travellingat constant speed through different routes. A rule of updating the probabilities is designed in orderto form constructive interference at target locations so that instantaneous energy can be maximised.Experiment is conducted on a 4-class classification problem extracted from MNIST. The results ex-hibit interesting but expected behavours, which do not exist in a bp updated network, but more likelearning in a real human, especially in the few-shot scenario.
Category: Artificial Intelligence
[1] viXra:2008.0130 [pdf] submitted on 2020-08-18 00:40:01
Authors: Vivek Verma
Comments: 2 Pages.
This paper describes the background and implementation behind a project that uses Neroevolution of Augmenting Topologies (NEAT) to play Super Mario Bros. It's implementation is different from classic applications of NEAT since the training process was heavily optimized using multithreading and downsampling. As a result, the training process can be run on underpowered CPUs without the help of an external GPU. The neural network successfully completed level 1-1 of the game.
Category: Artificial Intelligence
| Third-Party Links: |
|