| viXra.org > Artificial Intelligence > 2004 |
 |
 |
| viXra.org > Artificial Intelligence > 2004 |
|
|
[20] viXra:2004.0676 [pdf] submitted on 2020-04-29 17:09:39
Authors: Mohammed Hasan
Comments: 4 Pages.
Due to the rapid increase in the desire to use online technology, the use of password security has become vital for users worldwide to protect their sensitive data or accounts by implementing a password key only known to them in order to access their personal data. Throughout the years, as data has become more involved with being stored online, the creativity of different strategies of passwords has also increased as certain data may only be accessed through unique methods such as fingerprint scan. One of the major types of services that require users to protect their details is online banking such as PayPal or NatWest where users would provide a stronger password compared to an account with low value such as a mobile phone game. This report will go in depth on the best practices and strategies that derive from password security.
Category: Artificial Intelligence
[19] viXra:2004.0675 [pdf] submitted on 2020-04-29 19:10:40
Authors: Shashank Jain, Amritesh Singh, Rahul Ranjan Singh
Comments: 7 Pages.
There are many types of invoice having table
exist in the current system such as table in native text invoices, table in image invoices (II), table in handwritten invoices (HI) and so on. Nowadays, these different types of invoices are processing manually. Now our aim to survey
such system which can handle invoices having the table automatically by using OCR (Optical Character Reader) and Deep Learning Technologies. Moreover, we will also discussed multiple technologies and suggest the best model as per our survey
Category: Artificial Intelligence
[18] viXra:2004.0611 [pdf] replaced on 2021-11-24 05:55:15
Authors: Dimiter Dobrev
Comments: 62 Pages. Bulgarian language
We will reduce the task of creating AI to the task of finding an appropriate language for description of the world. This will not be a programing language because programing languages describe only computable functions, while our language will describe a somewhat broader class of functions. Another specificity of this language will be that the description will consist of separate modules. This will enable us look for the description of the world automatically such that we discover it module after module. Our approach to the creation of this new language will be to start with a particular world and write the description of that particular world. The point is that the language which can describe this particular world will be appropriate for describing any world.
Category: Artificial Intelligence
[17] viXra:2004.0580 [pdf] submitted on 2020-04-25 11:56:48
Authors: Satya Narayana
Comments: 3 Pages.
As everyone knows that Sentimental analysis plays an important role in these days because many start-ups have started with user-driven content [1]. Only finding the voice is not be the real time scenario so finding the Sentiment analysis of agent and customer separately is an important research area in natural language processing. Natural language processing has a wide range of applications like voice recognition, machine translation, product review, aspect-oriented product analysis, sentiment analysis and text classification etc [2]. This process will improve the business by analyze the emotions of the conversation with respect to the customer voice separately and also agent voice separately. In this project author going to perform speaker identification and analyze the sentiment of the customer and agent separately using Amazon Comprehend. Amazon Comprehend is a natural language processing (NLP) service that uses machine learning to extract the content of the voice. By using the speaker identification author can extract the unstructured data like images, voice etc separately so it is easy to analyze the business performance. Thus, will identify the emotions of the conversation and give the output whether the customer conversation is Positive, Negative, Neutral, or Mixed. To perform this author going to use some services from Aws due to some advantages like scaling the resources is easy compare to the normal process like doing physically such as support vector machine (SVM). AWS services like s3 is a object data store, Transcribe which generate the audio to text in raw format, Aws Glue is a ETL Service which will extract transform and load the data from the S3, Aws Comprehend is a NLP service used for finding sentiment of audio, Lambda is a server less where author can write a code, Aws Athena is a analyzing tools which will make complex queries in less time and last there is quick sight is a business intelligent tool where author can visualize the data of customers and also agents.
Category: Artificial Intelligence
[16] viXra:2004.0559 [pdf] submitted on 2020-04-23 19:33:20
Authors: Rajeev Kumar, Rajesh Budihal
Comments: 20 Pages.
The purpose of this research paper, the topic of credit card fraud detection has gained and developed fraudsters are increasing day by day among researches because of their frequent look in varied and widespread application within the field of various branches of information technology and engineering. For example, genetic algorithms, Behavior-based techniques, and Hidden Marks models are also used to address these problems of technology. Credit card fraud detection models for transactions are tested individually and proceed to whatever is most effective. This thesis aims to detect fraudulent transactions and develop some method of generating test data. These algorithms are a predictive approach in solving high complexity computational problems. We discussed a new method to goal or deal with detect fraud by filtering the above techniques to induce an improved result. These algorithms are a predictive approach in solving high complexity computational problems. It is an adaptation technique and evolutionary discovery that supports the existence of genetic and fittest. Implementation of efficient credit card fraud detection systems is mandatory for all credit card issuing companies or their customers to reduce their losses.
Category: Artificial Intelligence
[15] viXra:2004.0412 [pdf] submitted on 2020-04-17 08:01:03
Authors: George Rajna
Comments: 45 Pages.
Artificial intelligence (AI) can diagnose COVID-19 from CT scans, researchers in China claim [26]
Researchers in Berlin and Heidelberg have now developed an intelligent neural network that can predict the functions of proteins in the human body. [25]
AI combined with stem cells promises a faster approach to disease prevention. Andrew Masterson reports. [24]
According to product chief Trystan Upstill, the news app "uses the best of artificial intelligence to find the best of human intelligence—the great reporting done by journalists around the globe." [23]
Category: Artificial Intelligence
[14] viXra:2004.0371 [pdf] replaced on 2020-05-07 20:44:02
Authors: Jeongik Cho
Comments: 12 Pages.
Traditional deep neural network classifier receives input data and passes through hidden layers to output predicted labels. In this paper, I propose an Inverted Conditional Generator Classifier that uses conditional generators to find a pair of condition vector and latent vector that can generate the data closest to the input data, and predict the label of the input data.
The conditional generator is a generative model that receives latent vector and condition vector, and generates data with desired conditions. A decoder of conditional VAE [1] or a generator of conditional GAN [2] can be a conditional generator.
The inverted Conditional Generator Classifier uses a trained conditional generator as it is.
The inverted conditional generator classifier repeatedly performs gradient descent by taking the latent vector for each condition as a variable and the model parameter as a constant to find the data closest to the input data. Then, among the data generated for each condition, the condition vector of the data closest to the input data becomes the predicted label.
Inverted Conditional Generator Classifier is slow to predict because prediction is based on gradient descent, but has high accuracy and is very robust against adversarial attacks [3] such as noise.
In addition, the Inverted Conditional Generator Classifier can measure the degree of out-of-class through the difference between the generated nearest data and input data. A high degree of out-of-class means that the input data is separate from the cluster of each class, or Inverted Conditional Generator Classifier has little confidence in prediction. Through this, Inverted Conditional Generator Classifier can classify the input data as out-of-class or defer classification due to the lack of confidence in prediction.
Category: Artificial Intelligence
[13] viXra:2004.0363 [pdf] submitted on 2020-04-15 08:02:01
Authors: Amine Amyar, Romain Modzelewski, Su Ruan
Comments: 7 Pages.
The fast spreading of the novel coronavirus COVID-19 has aroused worldwide interest and concern, and caused more than one million and a half confirmed cases to date. To combat this spread, medical imaging such as computed tomography (CT) images can be used for diagnostic. An automatic detection tools is necessary for helping screening COVID-19 pneumonia using chest CT imaging. In this work, we propose a multitask deep learning model to jointly identify COVID-19 patient and segment COVID-19 lesion from chest CT images. Our motivation is to leverage useful information contained in multiple related tasks to help improve both segmentation and classification performances. Our architecture is composed by an encoder and two decoders for reconstruction and segmentation, and a multi-layer perceptron for classification. The proposed model is evaluated and compared with other image segmentation and classification techniques using a dataset of 1044 patients including 449 patients with COVID-19, 100 normal ones, 98 with lung cancer and 397 of different kinds of pathology. The obtained results show very encouraging performance of our method with a dice coefficient higher than 0.78 for the segmentation and an area under the ROC curve higher than 93% for the classification.
Category: Artificial Intelligence
[12] viXra:2004.0318 [pdf] submitted on 2020-04-12 21:21:53
Authors: Yuan Gao
Comments: 11 Pages.
One-stage object detectors like SSD and YOLO are able to speed up existing two-stage detectors like Faster R-CNN by removing the object proposal stage and making up for the lost performance in other ways. Nonetheless, the same approach is not easily transferable to instance segmentation task. Current one-stage instance segmentation methods can be simply classified into segmentation-based methods which segment first then do clustering, and proposal-based methods which detect first then predict masks for each instance proposal. Proposal-based methods always enjoy a better mAP; by contrast, segmentation-based methods are generally faster when inferencing. In this work, we first propose a one-stage segmentation-based instance segmentation solution, in which a pull loss and a push loss are used for differentiating instances. We then propose two post-processing methods, which provide a trade-off between accuracy and speed.
Category: Artificial Intelligence
[11] viXra:2004.0293 [pdf] submitted on 2020-04-11 22:56:08
Authors: Abhishek.B.N
Comments: 4 Pages.
Security algorithms enables secure communication between two parties in the presence of a third-Party or a snooper. It guarantees the recipient of the message of the genuineness of the received message, protects the message against the unauthorized release of the message content by the third party, only authorized users can access the data. MD5 and (SHA), cryptographic hash algorithms are one-way hashing functions which are easier to compute/convert but are much harder to reverse and would take around millions of years to compute the authentic message content. This research paper analyses the two hash algorithms, MD5 and SHA, using various key features. Their features have also been highlighted in order to provide a better comparison picture so that they can understand which algorithm has superseded the other.
Category: Artificial Intelligence
[10] viXra:2004.0248 [pdf] submitted on 2020-04-10 16:17:02
Authors: Rajdeep Singh
Comments: 3 Pages.
The novel coronavirus - COVID-19 - has evolved into a global pandemic. With that, it is imperative that countries and medical facilities are equipped with the technology and resources to give every person the greatest chance of surviving. With that, even developed nations are beginning to run low on medical supplies such as hospital beds, masks, and respirators. With the growth of cases in the United States, hospitals will continue to run out of supplies. It is imperative that medical supplies get distributed to those who need it the most first. This paper outlines a machine learning approach to predicting patients who are at the most risk of mortality given the confirmed positive diagnosis of coronavirus. The final results were inconclusive enough to be implemented in a real-world scenario.
Category: Artificial Intelligence
[9] viXra:2004.0222 [pdf] replaced on 2021-03-15 16:08:56
Authors: Xuezhe Ma, Xiang Kong, Shanghang Zhang, Eduard Hovy
Comments: 23 Pages. Published in ICLR 2021
In this work, we propose a new generative model that is capable of automatically decoupling global and local representations of images in an entirely unsupervised setting, by embedding a generative flow in the VAE framework to model the decoder. Specifically, the proposed model utilizes the variational auto-encoding framework to learn a (low-dimensional) vector of latent variables to capture the global information of an image, which is fed as a conditional input to a flow-based invertible decoder with architecture borrowed from style transfer literature. Experimental results on standard image benchmarks demonstrate the effectiveness of our model in terms of density estimation, image generation and unsupervised representation learning. Importantly, this work demonstrates that with only architectural inductive biases, a generative model with a likelihood-based objective is capable of learning decoupled representations, requiring no explicit supervision. The code for our model is available at https://github.com/XuezheMax/wolf.
Category: Artificial Intelligence
[8] viXra:2004.0190 [pdf] submitted on 2020-04-08 01:37:34
Authors: George Rajna
Comments: 41 Pages.
A team of researchers at Google's DeepMind has developed an AI system that is able to predict the movement of glass molecules as the material transitions between liquid and solid states. [25] A research team centered at Osaka University, in collaboration with RIKEN, has developed a system that can overcome these difficulties by automatically searching for, focusing on, imaging, and tracking single molecules within living cells. [24] But researchers at Purdue University are working on a solution, combining quantum algorithms with classical computing on small-scale quantum computers to speed up database accessibility. [23] Researchers at the University of Twente, working with colleagues at the Technical Universities of Delft and Eindhoven, have successfully developed a new and interesting building block. [22] Researchers at the Institut d'Optique Graduate School at the CNRS and Université Paris-Saclay in France have used a laser-based technique to rearrange cold atoms one-by-one into fully ordered 3D patterns. [21] Reduced entropy in a three-dimensional lattice of super-cooled, laser-trapped atoms could help speed progress toward creating quantum computers. [20] Under certain conditions, an atom can cause other atoms to emit a flash of light. At TU Wien (Vienna), this quantum effect has now been measured. [19] A recent discovery by William & Mary and University of Michigan researchers transforms our understanding of one of the most important laws of modern physics. [18] Now, a team of physicists from The University of Queensland and the NÉEL Institute has shown that, as far as quantum physics is concerned, the chicken and the egg can both come first. [17]
Category: Artificial Intelligence
[7] viXra:2004.0159 [pdf] submitted on 2020-04-07 03:45:06
Authors: Yang Zhang
Comments: 14 Pages.
Nature is structural instead of random, correlation is just approximation of causality, and data is not science: the more we reveal the more we revere nature on our voyage of unprecedented discovery. We argue that the soul(s) or exotic soul(s) of quotient Hypercomplex arbifold multiscale Spacetime (HyperSpacetime)'s corresponding manifold(s)/general (quotient and non-quotient) HyperSpacetime is the origin of super/general intelligence, and the metric of super/general intelligence is the complexity of quotient/general HyperSpacetime's corresponding generic polynomial. We also argue that the intersecting soul(s) and/or exotic soul(s) as varieties of quotient HyperSpacetime's corresponding manifold(s), when their maximal/minimum sectional curvatures approaching positive infinity and/or negative infinity as singularities, is the origin of quantum entanglement. We further argue
that the maximal/minimum sectional curvatures of the same intersecting soul(s) and/or exotic soul(s),
is the origin of convergent evolution through conformal transformation. We derive even N-dimensional HyperSpacetime, a M-open (\begin{math} M = C_{_{I+N}}^{^I} \text{, } I, N, M \to \infty \end{math})
arbifold as generalized orbifold with the structure of a algebraic variety $\mathcal{A}$, without or with loop group action as $\mathcal{A}=[\mathcal{M}/\mathcal{LG}]$ ($\mathcal{M}$ as complex manifold, $\mathcal{LG}$ as loop group), it arises from I-degree (power of 2) hypercomplex even N-degree generic polynomial continuous/discrete function/functor as nonlinear action functional in hypercomplex $\mathbb{HC}^{\infty}$ useful for generic neural networks: $\mathcal{F}(S_j,T_j)=\prod_{n=1}^{^{N}}(w_nS_n(T_n)+b_n+ \gamma \sum_{k=1}^{^{j}}\mathcal{F}(S_{k-1},T_{k-1}))$ where $j=1,\dots,N$, $S_{i}=s_0e_0+\sum_{i=1}^{^{{I-1}}}s_{i}e_{i}$, $T_{i}=t_0e_0+\sum_{i=1}^{^{{I-1}}}t_{i}e_{i}$ over noncommutative nonassociative loop group. Its sectional curvature is \begin{math}
\kappa = \frac{{\left| {\mathcal{F}''\left(X \right)} \right|}}{{{{\left( {1 + {{\left[ {\mathcal{F}'\left(X \right)} \right]}^2}} \right)}^{\frac{3}{2}}}}} \end{math} if $\mathcal{F}(X)$ is smooth, or \begin{math} \kappa = \kappa_{max}\kappa_{min}
\end{math} if nonsmooth, by correlating general relativity with quantum mechanics via extension from 3+1 dimensional spacetime $\mathbb{R}^{4}$ to even N-dimensional HyperSpacetime $\mathbb{HC}^{\infty}$. By directly addressing multiscale, singularities, statefulness, nonlinearity instead of via activation function and backpropagation, HyperSpacetime with its corresponding generic polynomial determining the complexity of ANN, rigorously models curvature-based $2^{nd}$ order optimization in arbifold-equivalent neural networks beyond gradient-based $1^{st}$ order optimization in manifold-approximated adopted in AI. We establish HyperSpacetime generic equivalence theory by synthesizing Generalized Poincar\'{e} conjecture, soul theorem, Galois theory, Fermat's last theorem, Riemann hypothesis, Hodge conjecture, Euler's theorem, Euclid theorem and universal approximation theorem. Our theory qualitatively and quantitatively tackles the black box puzzle in AI, quantum entanglement and convergent evolution. Our future work includes HyperSpacetime refinement, complexity reduction and synthesis as our ongoing multiversal endeavor.
Category: Artificial Intelligence
[6] viXra:2004.0106 [pdf] submitted on 2020-04-05 05:57:59
Authors: George Rajna
Comments: 51 Pages.
To this end, Ph.D. researcher Lars Banko, together with colleagues from the Interdisciplinary Centre for Advanced Materials Simulation at RUB, Icams for short, modified a so-called generative model. [30] Now, researchers have tested the first artificial intelligence model to identify and rank many causes in real-world problems without time-sequenced data, using a multi-nodal causal structure and Directed Acyclic Graphs. [29] A country that thinks its adversaries have or will get AI weapons will want to get them too. Wide use of AI-powered cyberattacks may still be some time away. [28] Following the old saying that "knowledge is power", companies are seeking to infer increasingly intimate properties about their customers as a way to gain an edge over their competitors. [27] Researchers from Human Longevity, Inc. (HLI) have published a study in which individual faces and other physical traits were predicted using whole genome sequencing data and machine learning. [26] Artificial intelligence can improve health care by analyzing data from apps, smartphones and wearable technology. [25] Now, researchers at Google's DeepMind have developed a simple algorithm to handle such reasoning-and it has already beaten humans at a complex image comprehension test. [24] A marimba-playing robot with four arms and eight sticks is writing and playing its own compositions in a lab at the Georgia Institute of Technology. The pieces are generated using artificial intelligence and deep learning. [23] Now, a team of researchers at MIT and elsewhere has developed a new approach to such computations, using light instead of electricity, which they say could vastly improve the speed and efficiency of certain deep learning computations. [22] Physicists have found that the structure of certain types of quantum learning algorithms is very similar to their classical counterparts-a finding that will help scientists further develop the quantum versions. [21] We should remain optimistic that quantum computing and AI will continue to improve our lives, but we also should continue to hold companies, organizations, and governments accountable for how our private data is used, as well as the technology's impact on the environment. [20]
Category: Artificial Intelligence
[5] viXra:2004.0083 [pdf] submitted on 2020-04-04 01:50:26
Authors: Nirmal Tej Kumar
Comments: 3 Pages. Short Communication & Technical Notes
A Technical Communication on Understanding & Exploring [ Recommender Systems + Machine Learning(ML) + NLP +QRNG/mruby+SmartDevices+IoT/HPC-High Performance Computing ] in the Context of Advanced Scientific Imaging Algorithms towards Software R&D Using Ruby –> [ Designing + Developing + Testing ] Heterogeneous Computing Environments.
{ https://www.semanticscholar.org/ - COVID 19 Information is our inspiration } ----- →
Category: Artificial Intelligence
[4] viXra:2004.0034 [pdf] submitted on 2020-04-02 04:56:48
Authors: George Rajna
Comments: 75 Pages.
Artificial intelligence (AI) may soon have a central role to play in the global battle against COVID-19. [42]
Simon Fraser University researchers will use their pioneering imaging technology—called Mango, for its bright colour— to develop coronavirus testing kits. [41]
According to the Centers for Disease Control and Prevention, common human coronaviruses usually cause mild to moderate upper-respiratory tract illnesses, like the common cold. [40]
Category: Artificial Intelligence
[3] viXra:2004.0029 [pdf] submitted on 2020-04-02 08:48:20
Authors: George Rajna
Comments: 26 Pages.
For the first time, a team at the US Department of Energy's (DOE's) Oak Ridge National Laboratory (ORNL) is using artificial intelligence (AI) to find patterns in neutron scattering data that can lead to an understanding of the physics inside quantum or complex magnetic materials. [15] "As far as we know, this is the first published work showing an application of super resolution to neutrons. We're at the forefront of an exciting new trend that will help other neutron scattering facilities improve their own data resolution as well," said Lin. [14] Coupled with SNS, the world's most powerful pulsed accelerator-based neutron source, VENUS will be the only open research facility platform in the US to provide time-of-flight neutron imaging capabilities to users from academia and industry. [13] A spallation neutron source has been used by physicists in Japan to search for possible violations of the inverse square law of gravity. [12] Physicists have proposed a way to test quantum gravity that, in principle, could be performed by a laser-based, table-top experiment using currently available technology. [11] Now however, a new type of materials, the so-called Weyl semimetals, similar to 3-D graphene, allow us to put the symmetry destructing quantum anomaly to work in everyday phenomena, such as the creation of electric current. [10] Physicist Professor Chunnong Zhao and his recent PhD students Haixing Miao and Yiqiu Ma are members of an international team that has created a particularly exciting new design for gravitational wave detectors. [9] A proposal for a gravitational-wave detector made of two space-based atomic clocks has been unveiled by physicists in the US. [8] The gravitational waves were detected by both of the twin Laser Interferometer Gravitational-Wave Observatory (LIGO) detectors, located in Livingston, Louisiana, and Hanford, Washington, USA. [7] A team of researchers with the University of Lisbon has created simulations that indicate that the gravitational waves detected by researchers with the LIGO project, and which are believed to have come about due to two black holes colliding, could just have easily come from another object such as a gravaster (objects which are believed to have their insides made of dark energy) or even a wormhole. In their paper published in Physical Review Letters, the team describes the simulations they created, what was seen and what they are hoping to find in the future. [6] In a landmark discovery for physics and astronomy, international scientists said Thursday they have glimpsed the first direct evidence of gravitational waves, or ripples in space-time, which Albert Einstein predicted a century ago. [5] Scientists at the National Institute for Space Research in Brazil say an undiscovered type of matter could be found in neutron stars (illustration shown). Here matter is so dense that it could be 'squashed' into strange matter. This would create an entire 'strange star'-unlike anything we have seen. [4] The changing acceleration of the electrons explains the created negative electric field of the magnetic induction, the electromagnetic inertia, the changing relativistic mass and the Gravitational Force, giving a Unified Theory of the physical forces. Taking into account the Planck Distribution Law of the electromagnetic oscillators also, we can explain the electron/proton mass rate and the Weak and Strong Interactions.
Category: Artificial Intelligence
[2] viXra:2004.0024 [pdf] submitted on 2020-04-01 10:54:54
Authors: George Rajna
Comments: 39 Pages.
Researchers at the Institute of Industrial Science, a part of The University of Tokyo, demonstrated a novel artificial intelligence system that can find and label 2-D materials in microscope images in the blink of an eye. [24] The research group took advantage of a system at SLAC's Stanford Synchrotron Radiation Lightsource (SSRL) that combines machine learning-a form of artificial intelligence where computer algorithms glean knowledge from enormous amounts of data-with experiments that quickly make and screen hundreds of sample materials at a time. [23] Researchers at the UCLA Samueli School of Engineering have demonstrated that deep learning, a powerful form of artificial intelligence, can discern and enhance microscopic details in photos taken by smartphones. [22] Such are the big questions behind one of the new projects underway at the MIT-IBM Watson AI Laboratory, a collaboration for research on the frontiers of artificial intelligence. [21]
Category: Artificial Intelligence
[1] viXra:2004.0003 [pdf] submitted on 2020-04-01 09:41:22
Authors: George Rajna
Comments: 40 Pages.
Recently, a research team from Shanghai Institute of Optics and Fine Mechanics of the Chinese Academy of Sciences (CAS) proposed a three-dimensional damage localization method which was insensitive to the type of damage. [26]
A UCLA research team has devised a technique that extends the capabilities of fluorescence microscopy, which allows scientists to precisely label parts of living cells and tissue with dyes that glow under special lighting. [25]
Social, economic, environmental and health inequalities within cities can be detected using street imagery. [24]
Category: Artificial Intelligence
| Third-Party Links: |
|