Downloads & Free Reading Options - Results

Programming by Example (PBE) targets at automatically inferring a computer program for accomplishing a certain task from sample input and output. In this paper, we propose a deep neural networks (DNN) based PBE model called Neural Programming by Example (NPBE), which can learn from input-output strings and induce programs that solve the string manipulation problems. Our NPBE model has four neural network based components: a string encoder, an input-output analyzer, a program generator, and a symbol selector. We demonstrate the effectiveness of NPBE by training it end-to-end to solve some common string manipulation problems in spreadsheet systems. The results show that our model can induce string manipulation programs effectively. Our work is one step towards teaching DNN to generate computer programs.

Programming by Example (PBE) targets at automatically inferring a computer program for accomplishing a certain task from sample input and output. In this paper, we propose a deep neural networks (DNN) based PBE model called Neural Programming by Example (NPBE), which can learn from input-output strings and induce programs that solve the string manipulation problems. Our NPBE model has four neural network based components: a string encoder, an input-output analyzer, a program generator, and a symbol selector. We demonstrate the effectiveness of NPBE by training it end-to-end to solve some common string manipulation problems in spreadsheet systems. The results show that our model can induce string manipulation programs effectively. Our work is one step towards teaching DNN to generate computer programs.

مجموعة.كتب.ومقالات.ومحاضرات.ورسائل.وأوراق.في.مجال.تطوير.الذات

Programming language processing (similar to natural language processing) is a hot research topic in the field of software engineering; it has also aroused growing interest in the artificial intelligence community. However, different from a natural language sentence, a program contains rich, explicit, and complicated structural information. Hence, traditional NLP models may be inappropriate for programs. In this paper, we propose a novel tree-based convolutional neural network (TBCNN) for programming language processing, in which a convolution kernel is designed over programs' abstract syntax trees to capture structural information. TBCNN is a generic architecture for programming language processing; our experiments show its effectiveness in two different program analysis tasks: classifying programs according to functionality, and detecting code snippets of certain patterns. TBCNN outperforms baseline methods, including several neural models for NLP.

This paper shows that a new type of artificial neural network (ANN) -- the Simultaneous Recurrent Network (SRN) -- can, if properly trained, solve a difficult function approximation problem which conventional ANNs -- either feedforward or Hebbian -- cannot. This problem, the problem of generalized maze navigation, is typical of problems which arise in building true intelligent control systems using neural networks. (Such systems are discussed in the chapter by Werbos in K.Pribram, Brain and Values, Erlbaum 1998.) The paper provides a general review of other types of recurrent networks and alternative training techniques, including a flowchart of the Error Critic training design, arguable the only plausible approach to explain how the brain adapts time-lagged recurrent systems in real-time. The C code of the test is appended. As in the first tests of backprop, the training here was slow, but there are ways to do better after more experience using this type of network.

[1] Introduction : Texture represents one of the four principal factors defining food/fruit quality, together with appearance, flavour and nutritional properties (Bourne, 2002), and plays a key role in consumer acceptability and recognition of quince. Textural characteristics of quinces defined by “crispness”, “juiciness”, “hardness”,“firmness” and “mealiness” are often key drivers of consumer preference. Many non-destructive methods, including image analysis, spectroscopy, ultrasound and sound techniques, have been developed to diagnose internal and external defects in fruits and vegetables. Cheng and Haugh (1994) used a frequency of 250-kHz, rather than 1-MHz, to detect hollow heart. They were not able to transmit successfully the ultrasound wave through the whole tuber using 1-MHz transducers but found the 250-kHz transducers to be practical for a transmission path length of up to 89.7 mm. In a research an acoustic setup was developed to simultaneously detect the resonant frequencies from equator and from calyx shoulder of pear. The researchers proposed index based on these two frequencies was used for firmness evaluation of non-spherical pear; Compared with two types of single frequency-based indices, the firmness sensitivity of the dual-frequency index is mostly close to that of MT penetration test. The firmness index can classify pears with a high total accuracy (93.4%), making it suitable for nondestructive detection of firmness of differently shaped pears (Zhang et al., 2018). The goal of this study was to develop a nondestructive method based on acoustic impulse response of quince fruit using genetic programming and artificial neural network during storage.   Materials and Methods : In the experiment 120 quince fruits ( Cydonia oblonga ) were harvested from a field near Isfahan 181 days after full flowering of the trees. For each cultivar, only samples of similar size and without visible external damage were chosen. The samples were packed in sterile nylon bags and stored at 4°C. Non-destructive test (acoustic response) as well as destructive test (chemical measurement and penetration test) were performed every 15 days for 4 months (Akbari Bisheh et al., 2014). Total soluble solids (TSS) were determined by a hand refractometer device (model: MT03 Japan) and expressed as °Brix. Ascorbic acid of the juice was measured by titration with copper sulfate and potassium iodide based on the Barakat et al. (1973) procedure. Titratable acidity was measured according to the AOAC method. To determine the total phenol content of juice, the Waterhouse method (2000) was used. Determination of the pH of the fruit extract using a pH meter (Portable Model P-755, Japan). Physical attributes of the samples including volume as well as major, minor, intermittent diameters and mass were calculated using the relations proposed by Stroshine and Hammand (1994). Penetration test was conducted by the material test machine (SANTAM, STM-20 model, Iran). In order to analyze the response sound signal of quince in time and frequency domain, a system equipped with a sample holder with foam rubber covered surface, an impact mechanism, a microphone and an electronic circuit was utilized. To record impact sound features a microphone was positioned next to the fruit and was hit at three speed level (0 . 3, 0.9 and 1.5 m/s). After recoding sound, five features (acoustic peak, maximum acoustic pressure, mean acoustic pressure and natural frequency) were extracted and used as inputs for models. In order to predict the stiffness, four methods of genetic programming, neural network and existing mathematical models (FI and SIQ-FT) were used. In order to carry out statistical analysis, analysis of variance (ANOVA) and Duncan's multiple range test at 5% probability level were performed according to the completely randomized design (CRD).   Results and discussion : In this study, Duncan's multiple range comparison test was used to investigate the significant difference between destructive and non-destructive parameters at 5% probability level. According to the results, acoustic peak, maximum acoustic pressure, mean acoustic pressure and natural frequency were decreased by increasing storage time. Statistical analysis of the destructive tests also showed a decreasing trend at the 5% level. In several papers, two mathematical equations have been used to obtain the relationship between the mass resonance frequency and the sound of impact. In this study, genetic programming and neural network modeling were used to compare the results of these relationships. The regression coefficients between the actual and the predicted values for the resonance-mass relation and the effect of the sound from the collision were R 2 = 0.601 and R 2 = 0.754, respectively. Also, the regression values obtained from genetic programming and neural network modeling were R 2 = 0.9567 and R 2 = 0.933, respectively. In a research, the overall R 2 value amounts for stiffness prediction was reported to be 0.79 (Schotte et al ., 1999). Abbaszadeh et al . (2013) evaluated watermelons texture using their vibration responses. They declared their proposed method could predict textural acceptability of watermelons with determination coefficients 0.99. According to the obtained values, the best methods for stiffness prediction were genetic programming and f neural network methods, respectively.  

This work investigates the application of evolutionary programming, a stochastic search technique, for simultaneously determining the weights and the number of hidden units in a fully-connected, multi-layer neural network. The simulated evolution search paradigm provides a means for optimizing both network structure and weight coefficients. Orthogonal learning is implemented by independently modifying network structure and weight parameters. Different structural level search strategies are investigated by comparing the training processes for the 3-bit parity problem. The results indicate that evolutionary programming provides a robust framework for evolving neural networks. Neural Networks, Evolutionary Programming, Signal Detection

xxxiv, 494 pages ; 19 cm

xxxiv, 494 pages ; 19 cm

We present a method for automatically generating repair feedback for syntax errors for introductory programming problems. Syntax errors constitute one of the largest classes of errors (34%) in our dataset of student submissions obtained from a MOOC course on edX. The previous techniques for generating automated feed- back on programming assignments have focused on functional correctness and style considerations of student programs. These techniques analyze the program AST of the program and then perform some dynamic and symbolic analyses to compute repair feedback. Unfortunately, it is not possible to generate ASTs for student pro- grams with syntax errors and therefore the previous feedback techniques are not applicable in repairing syntax errors. We present a technique for providing feedback on syntax errors that uses Recurrent neural networks (RNNs) to model syntactically valid token sequences. Our approach is inspired from the recent work on learning language models from Big Code (large code corpus). For a given programming assignment, we first learn an RNN to model all valid token sequences using the set of syntactically correct student submissions. Then, for a student submission with syntax errors, we query the learnt RNN model with the prefix to- ken sequence to predict token sequences that can fix the error by either replacing or inserting the predicted token sequence at the error location. We evaluate our technique on over 14, 000 student submissions with syntax errors. Our technique can completely re- pair 31.69% (4501/14203) of submissions with syntax errors and in addition partially correct 6.39% (908/14203) of the submissions.

477 pages 19 cm

This work investigates the application of evolutionary programming, a stochastic search technique, for simultaneously determining the weights and the number of hidden units in a fully-connected, multi-layer neural network. The simulated evolution search paradigm provides a means for optimizing both network structure and weight coefficients. Orthogonal learning is implemented by independently modifying network structure and weight parameters. Different structural level search strategies are investigated by comparing the training processes for the 3-bit parity problem. The results indicate that evolutionary programming provides a robust framework for evolving neural networks.

This work investigates the application of evolutionary programming, a stochastic search technique, for simultaneously determining the weights and the number of hidden units in a fully-connected, multi-layer neural network. The simulated evolution search paradigm provides a means for optimizing both network structure and weight coefficients. Orthogonal learning is implemented by independently modifying network structure and weight parameters. Different structural level search strategies are investigated by comparing the training processes for the 3-bit parity problem. The results indicate that evolutionary programming provides a robust framework for evolving neural networks.

This article aims to achieve two goals: to show that probability is not the only way of dealing with uncertainty (and even more, that there are kinds of uncertainty which are for principled reasons not addressable with probabilistic means); and to provide evidence that logic-based methods can well support reasoning with uncertainty. For the latter claim, two paradigmatic examples are presented: Logic Programming with Kleene semantics for modelling reasoning from information in a discourse, to an interpretation of the state of affairs of the intended model, and a neural-symbolic implementation of Input/Output logic for dealing with uncertainty in dynamic normative contexts.

Empirically, neural networks that attempt to learn programs from data have exhibited poor generalizability. Moreover, it has traditionally been difficult to reason about the behavior of these models beyond a certain level of input complexity. In order to address these issues, we propose augmenting neural architectures with a key abstraction: recursion. As an application, we implement recursion in the Neural Programmer-Interpreter framework on four tasks: grade-school addition, bubble sort, topological sort, and quicksort. We demonstrate superior generalizability and interpretability with small amounts of training data. Recursion divides the problem into smaller pieces and drastically reduces the domain of each neural network component, making it tractable to prove guarantees about the overall system's behavior. Our experience suggests that in order for neural architectures to robustly learn program semantics, it is necessary to incorporate a concept like recursion.

The human perception system is multi-dimensional; humans process more than just the sound of the word. Any speech recognition system that mimics human speech perception will need to be multi-dimensional. This methodology formed the basis for the design approach used in this research effort. Linear Predictive Coefficients (LPC) and formants were used as distinct and independent inputs into a recognition system consisting of a Kohonen neural network and a dynamic programming word classifier. A feature-fusion section and rule-based system were used to integrate the two input feature sets into one output result. This research effort involved extensive testing of the Kohonen network. Using a speech input signal, different Kohonen gain reduction methods, initial gain values, and conscience values were tested for various iteration times in an effort to quantify the response and capabilities of the Kohonen network. Three- dimensional Kohonen-Dynamic Programming surfaces were developed that graphically showed the effects of gain, conscience, and iteration time on the speech recognition response of a Kohonen neural network. A new standard iteration time called a multiple was used during training of the Kohonen networks. The results of the basic research on the Kohonen produced an optimized Kohonen configuration that was used in the multiple-feature recognition system. A 70-word vocabulary of F-16 cockpit commands were used to evaluate the new feature-fusion method. Theses

The human perception system is multi-dimensional; humans process more than just the sound of the word. Any speech recognition system that mimics human speech perception will need to be multi-dimensional. This methodology formed the basis for the design approach used in this research effort. Linear Predictive Coefficients (LPC) and formants were used as distinct and independent inputs into a recognition system consisting of a Kohonen neural network and a dynamic programming word classifier. A feature-fusion section and rule-based system were used to integrate the two input feature sets into one output result. This research effort involved extensive testing of the Kohonen network. Using a speech input signal, different Kohonen gain reduction methods, initial gain values, and conscience values were tested for various iteration times in an effort to quantify the response and capabilities of the Kohonen network. Three- dimensional Kohonen-Dynamic Programming surfaces were developed that graphically showed the effects of gain, conscience, and iteration time on the speech recognition response of a Kohonen neural network. A new standard iteration time called a multiple was used during training of the Kohonen networks. The results of the basic research on the Kohonen produced an optimized Kohonen configuration that was used in the multiple-feature recognition system. A 70-word vocabulary of F-16 cockpit commands were used to evaluate the new feature-fusion method. Theses

This work investigates the application of evolutionary programming, a stochastic search technique, for determining connectivity in feedforward neural networks. The method is capable of simultaneously evolving both the connection scheme and the network weights. The number of synapses are incorporated into an objective function so that network parameter optimization is done with respect to a connectivity cost as well as mean pattern error. Experimental results are shown using feedforward networks for simple binary mapping problems.

This article is from PLoS ONE , volume 9 . Abstract Neural stem/progenitor cells (NSC) have the potential for treatment of a wide range of neurological diseases such as Parkinson Disease and multiple sclerosis. Currently, NSC have been isolated only from hippocampus and subventricular zone (SVZ) of the adult brain. It is not known whether NSC can be found in all parts of the developing mid-trimester central nervous system (CNS) when the brain undergoes massive transformation and growth. Multipotent NSC from the mid-trimester cerebra, thalamus, SVZ, hippocampus, thalamus, cerebellum, brain stem and spinal cord can be derived and propagated as clonal neurospheres with increasing frequencies with increasing gestations. These NSC can undergo multi-lineage differentiation both in vitro and in vivo, and engraft in a developmental murine model. Regionally-derived NSC are phenotypically distinct, with hippocampal NSC having a significantly higher neurogenic potential (53.6%) over other sources (range of 0%–27.5%, p

Dataflow matrix machines arise naturally in the context of synchronous dataflow programming with linear streams. They can be viewed as a rather powerful generalization of recurrent neural networks. Similarly to recurrent neural networks, large classes of dataflow matrix machines are described by matrices of numbers, and therefore dataflow matrix machines can be synthesized by computing their matrices. At the same time, the evidence is fairly strong that dataflow matrix machines have sufficient expressive power to be a convenient general-purpose programming platform. Because of the network nature of this platform, programming patterns often correspond to patterns of connectivity in the generalized recurrent neural networks understood as programs. This paper explores a variety of such programming patterns.

We study self-programming in recurrent neural networks where both neurons (the `processors') and synaptic interactions (`the programme') evolve in time simultaneously, according to specific coupled stochastic equations. The interactions are divided into a hierarchy of $L$ groups with adiabatically separated and monotonically increasing time-scales, representing sub-routines of the system programme of decreasing volatility. We solve this model in equilibrium, assuming ergodicity at every level, and find as our replica-symmetric solution a formalism with a structure similar but not identical to Parisi's $L$-step replica symmetry breaking scheme. Apart from differences in details of the equations (due to the fact that here interactions, rather than spins, are grouped into clusters with different time-scales), in the present model the block sizes $m_i$ of the emerging ultrametric solution are not restricted to the interval $[0,1]$, but are independent control parameters, defined in terms of the noise strengths of the various levels in the hierarchy, which can take any value in $[0,\infty\ket$. This is shown to lead to extremely rich phase diagrams, with an abundance of first-order transitions especially when the level of stochasticity in the interaction dynamics is chosen to be low.

We study self-programming in recurrent neural networks where both neurons (the `processors') and synaptic interactions (`the programme') evolve in time simultaneously, according to specific coupled stochastic equations. The interactions are divided into a hierarchy of $L$ groups with adiabatically separated and monotonically increasing time-scales, representing sub-routines of the system programme of decreasing volatility. We solve this model in equilibrium, assuming ergodicity at every level, and find as our replica-symmetric solution a formalism with a structure similar but not identical to Parisi's $L$-step replica symmetry breaking scheme. Apart from differences in details of the equations (due to the fact that here interactions, rather than spins, are grouped into clusters with different time-scales), in the present model the block sizes $m_i$ of the emerging ultrametric solution are not restricted to the interval $[0,1]$, but are independent control parameters, defined in terms of the noise strengths of the various levels in the hierarchy, which can take any value in $[0,\infty\ket$. This is shown to lead to extremely rich phase diagrams, with an abundance of first-order transitions especially when the level of stochasticity in the interaction dynamics is chosen to be low.

For achieving success in a global arena cross cultural training should be provided to employees to settle down between the global business environment and culture as one of the factors contributing to economic success, revenue generation, surplus booking, goodwill enhancement, market fame and many more. More the revenue, more the profit booking leads to rise company’s goodwill and builds customers faith as well as provides employee satisfaction which motivates employees to be more productive, more efficient, more energetic, more enthusiastic, and never let employees to get stressed from their work.AI ANN and goal programming is being used a method to find something fruitful to mitigate cross cultural issues in an organization. Shefali G | Dr. Rajesh Singh "Assessment of Neural Network and Goal Programming on Cross Cultural Management" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-5 | Issue-4 , June 2021, URL: https://www.ijtsrd.compapers/ijtsrd41217.pdf Paper URL: https://www.ijtsrd.comcomputer-science/computer-network/41217/assessment-of-neural-network-and-goal-programming-on-cross-cultural-management/shefali-g

For achieving success in a global arena cross cultural training should be provided to employees to settle down between the global business environment and culture as one of the factors contributing to economic success, revenue generation, surplus booking, goodwill enhancement, market fame and many more. More the revenue, more the profit booking leads to rise company’s goodwill and builds customers faith as well as provides employee satisfaction which motivates employees to be more productive, more efficient, more energetic, more enthusiastic, and never let employees to get stressed from their work.AI ANN and goal programming is being used a method to find something fruitful to mitigate cross cultural issues in an organization. Shefali G | Dr. Rajesh Singh "Assessment of Neural Network and Goal Programming on Cross Cultural Management" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-5 | Issue-4 , June 2021, URL: https://www.ijtsrd.compapers/ijtsrd41217.pdf Paper URL: https://www.ijtsrd.comcomputer-science/computer-network/41217/assessment-of-neural-network-and-goal-programming-on-cross-cultural-management/shefali-g

Radial Basis Function Neural Network (RBFNN) is very prominent in data processing. However, improving this technique is vital for the NN training process. This paper presents an integrated 2 Satisfiability in radial basis function neural network (RBFNN-2SAT). There are two different types of training in RBFNN, namely no-training technique and half-training technique. The performance of the solutions via Genetic Algorithm (GA) training was investigated by comparing the Radial Basis Function Neural Network No-Training Technique (RBFNN- 2SATNT) and Radial Basis Function Neural Network Half-Training Technique (RBFNN- 2SATHT). The comparison of both techniques was examined on 2 Satisfiability problem by using a C# software that was developed for this experiment. The performance of the RBFNN-2SATNT and RBFNN-2SATHT in performing 2SAT is discussed in terms of root mean squared error (RMSE), sum squared error (SSE), mean absolute percentage error (MAPE), mean absolute error (MAE), number of the hidden neurons and CPU time. Results obtained from a computer simulation showed that RBFNN-2SATHT outperformed RBFNN-2SATNT. 

This article describes the training of artificial neural network. A program for training a single – layer neural network is created and the result is shown.