Automatic Generation Research Articles

TyBox: An Automatic Design and Code Generation Toolbox for TinyML Incremental On-Device Learning

Incremental on-device learning is one of the most relevant and interesting challenges in the field of Tiny Machine Learning (TinyML). Indeed, differently from traditional TinyML solutions where the training is typically carried out on the Cloud and inference only occurs on the tiny devices (e.g., embedded systems or Internet-of-Things units), incremental on-device TinyML allows both the inference and the training of TinyML models directly on tiny devices. This ability paves the way for TinyML-enabled intelligent devices that can learn directly on the field and adapt to evolving environments, different working conditions, or specific users. The literature in this field is quite limited with very few solutions focusing only on the incremental fine-tuning of machine learning models, whereas a general solution encompassing algorithms and code generation for incremental on-device TinyML is still perceived as missing. The aim of this article is to introduce, to the best of our knowledge for the first time in the literature, a toolbox called TyBox for the automatic design and code generation of incremental on-device TinyML classification models. In more detail, starting from a “static” TinyML model, TyBox is able to (i) automatically design the “incremental” on-device version of the TinyML model that has been suitably designed to take into account the technological constraint on the RAM memory of the target tiny device, and (ii) autonomously provide the C++ codes and libraries to support the inference and learning of the incremental on-device TinyML model directly on the tiny devices. TyBox has been extensively compared with a state-of-the-art incremental learning solution for TinyML and tested on an off-the-shelf tiny device (i.e., the Arduino Nano 33 BLE) in three relevant TinyML application tasks and scenarios: binary image classification, multi-class image classification, and ultra-wide-band human activity recognition. In addition, TyBox is released to the scientific community as a public repository.

Automatic Reconstruction of 3D Models from 2D Drawings: A State-of-the-Art Review

Among the methods of 3D reconstruction, the automatic generation of 3D models from building documentation is one of the most accessible and inexpensive. For 30 years, researchers have proposed multiple methods to automatically generate 3D models from 2D drawings. This study compiles this research and discusses the different methods used to generate 3D models from 2D drawings. It offers a critical review of these methods, focusing on the coverage and completeness of the reconstruction process. This review allows us to identify the research gaps in the literature, and opportunities for improvement are identified for future research.

Novel Feature-Based Difficulty Prediction Method for Mathematics Items Using XGBoost-Based SHAP Model

The level of difficulty of mathematical test items is a critical aspect for evaluating test quality and educational outcomes. Accurately predicting item difficulty during test creation is thus significantly important for producing effective test papers. This study used more than ten years of content and score data from China’s Henan Provincial College Entrance Examination in Mathematics as an evaluation criterion for test difficulty, and all data were obtained from the Henan Provincial Department of Education. Based on the framework established by the National Center for Education Statistics (NCES) for test item assessment methodology, this paper proposes a new framework containing eight features considering the uniqueness of mathematics. Next, this paper proposes an XGBoost-based SHAP model for analyzing the difficulty of mathematics tests. By coupling the XGBoost method with the SHAP method, the model not only evaluates the difficulty of mathematics tests but also analyzes the contribution of specific features to item difficulty, thereby increasing transparency and mitigating the “black box” nature of machine learning models. The model has a high prediction accuracy of 0.99 for the training set and 0.806 for the test set. With the model, we found that parameter-level features and reasoning-level features are significant factors influencing the difficulty of subjective items in the exam. In addition, we divided senior secondary mathematics knowledge into nine units based on Chinese curriculum standards and found significant differences in the distribution of the eight features across these different knowledge units, which can help teachers place different emphasis on different units during the teaching process. In summary, our proposed approach significantly improves the accuracy of item difficulty prediction, which is crucial for intelligent educational applications such as knowledge tracking, automatic test item generation, and intelligent paper generation. These results provide tools that are better aligned with and responsive to students’ learning needs, thus effectively informing educational practice.

AI Based IT Training System

Retaining learner engagement is a major challenge in online learning environments, which is even more intensified with learning spaces increasingly built by combining resources from multiple independent sources. Narrative-centric learning experience has been found to improve learner engagement by several researchers. Towards this end, we propose an AI-based approach that generates auxiliary learning content called narrative fragments which are interspersed into the learning pathways to create interactive learning narratives. The proposed approach consists of the automatic generation of two types of narrative fragments– overviews of the learning pathway segments and reflection quizzes or formative assessments from learning resources in any format including open educational resources. The pipeline for the generation of the narrative fragments consists of various components based on different semantic models and a natural language generation (NLG) component based on a pre- trained language model GPT-2 (Generative Pre-trained Transformer 2). Automation enables the generation of narrative fragments on the fly whenever there are changes in the learning pathway due to the need for reiteration of concepts, pre-requisite knowledge acquisition, etc., enabling adaptability in the learning pathways. The proposed approach is domain agnostic which makes it easily adaptable to different domains. The NLG model is evaluated using ROUGE scores against several baselines. Automatically generated narrative fragments are evaluated by human evaluators. We obtained encouraging results in both cases

Comparative Study on Automated Reference Summary Generation using BERT Models and ROUGE Score Assessment

Automatic text summarization is a sub-area in text mining in which a computer system determines the most informative information in the original text to produce a summary for certain jobs and users. In the development of the systems, one of the most important tasks is to evaluate the quality of summaries produced by the systems. Generally, the evaluation task becomes laborious, time-consuming, and expensive because it requires significant efforts on annotation tasks for humans to manually create reference summaries. Being able to generate automatic reference summaries would promote the development of summarization systems in term of speed and evaluation. In this paper, we proposed an Auto-Ref Summary Generation framework for automatically generating reference summaries used in the generic text summarization evaluation task, the Sliced Summary. Given a set of clusters from a cluster ground-truth label dataset, variants of BERT models were utilized for creating cluster representations. The automatic reference summaries were later generated through a centroid-based summarization approach. Overall, DistilBERT, ROBERTa, and SBERT have played crucial roles in automatic summary generation, achieving the highest ROUGE-1 score of 0.47060. However, this does not meet our expectation on text coherence and readability aspects. Although the summaries generated through our proposed framework could not be used as the replacement of the manual summaries, this study has shed new light on the acquisition of automatic reference summaries from a ground-truth label dataset.

Frequency Control of Interconnected Power System via SBO assisted Simultaneous Tuning Strategy in Presence of DFIG

To achieve the objectives of the automatic generation control (AGC) of wind integrated interconnected power system (WIIPS), better tuning and control strategies are required. This paper suggests simultaneous tuning strategy for the AGC and wind system controllers to improve the stability of the WIIPS. The bulk amount of wind is integrated into the grid using the doubly fed induction generator (DFIG) and its simplified model is used to simulate the WIIPS. Initially, classical controllers are opted in AGC mechanism to reduce the frequency disturbances caused by the load and wind generation uncertainties. Further, classical controllers of WIIPS are replaced with cascade controllers to enhance the system frequency profile during the disturbances. The optimal set of solutions of the classical and cascade controllers are examined by the satin bowerbird optimizer (SBO) with the help of integral square error as fitness function of the optimization problem. Comparisons are provided to show the merits of the WIIPS over normal interconnected power system, cascade controller over classical controller and simultaneous tuning strategy over regular tuning mechanism.

Joint non-fragile automatic generation control and multi-event driven mechanism co-design for wind integrated power system under denial of services attacks

Denial of services (DoS) attacks exist in wind integrated power system. DoS attacks can cause network-induced delay and packages loss in information transmission. Meanwhile, considering the parameter perturbation of controller and system model uncertainty in wind integrated power system, these may cause the system dynamic performances degradation or even instability. Based on the above considerations, the joint non-fragile automatic generation robust control of wind integrated power system under DoS attacks is studied in this paper. In order to ensure the expected system performance and more effectively utilize the limited network communication resources under DoS attacks, a novel dynamic multi-event driven mechanism based joint non-fragile H∞ automatic generation control method is proposed. By constructing a suitable Lyapunov-Krasovskii functional and utilizing the Shur complement lemma to handle nonlinear matrix inequality, the sufficient conditions are derived to guarantee the asymptotic stability of wind integrated power system under DoS attacks. Furthermore, the performance of the proposed non-fragile regulator is demonstrated through a four-area wind integrated power system to show the feasibility and applicability. The analysis result indicates that the proposed scheme provides stronger robustness, higher wind energy utilization efficiency and more efficient communication mechanism.

Automatic bi-modal question title generation for Stack Overflow with prompt learning

Automatic bi-modal question title generation for Stack Overflow with prompt learning

Microstructure generation and full-field multi-scale analyses for short fiber reinforced thermoplastics: Application to PA66GF composites

Short Fiber Reinfored Thermoplastics (SFRTs) like PA66GF composites are highly heterogeneous materials with a rather complex microstructure such that the characterization and the prediction of their mechanical behavior remain quite challenging. So far, most of the research efforts have employed phenomenologigal and mean-field multi-scale models to deal with SFRTs. The present contribution rather focuses on a full-field multi-scale approach that incorporates advanced techniques in terms of microstructural representation and material modeling, allowing a deep insight of the dominating deformation mechanisms occurring in PA66GF composites. The proposed approach is based on an automatic periodic mesh generation algorithm for matrix-inclusion Representative Volume Elements (RVE) with randomly positioned fibers that follow a given Orientation Distribution Function (ODF). At the microscopic scale, while the fibers are assumed to be elastic, the behavior of the thermoplastic matrix is described by a phenomenological multi-mechanism constitutive model accounting for viscoelasticity, viscoplasticity and ductile damage. It results an advanced multi-scale model that enables to visualize the local deformation and degradation mechanisms occurring at the microscopic scale while simultaneously analyzing their influence on the macroscopic response of the composite upon monotonic, persistent and cyclic loading. The potential of the proposed approach is further evaluated by comparing the predicted responses against experimental data.

Accurate identification and confidence evaluation of automatic generation control command execution effect based on deep learning fusion model

With the increasing complexity of the power grid, the precision of the thermal power units' execution of automatic generation control (AGC) commands is gradually increasing the impact on the online regulation of the power grid. The deviation between the actual output of thermal power units and the AGC command of the grid will not only affect the consumption of new energy output, but also endanger the safe operation of the grid. This paper introduces “deep learning” technology to solve the problem. Firstly, an AGC command execution effect identification and confidence evaluation algorithm (ACEEI-CEA) is proposed. The algorithm builds a neural network model to accurately predict the unit output and the confidence evaluation of the prediction results. Next, a high-dimensional input preprocessing strategy based on variational autoencoder (VAE) is proposed to reduce the dimensionality of the model input attributes, improving the convergence and accuracy of the model. Finally, an AGC optimal command fast inversion solution method (AOCFISM) is designed. This method transforms the unit output deviation problem into an objective optimisation problem. And improve the efficiency of solving the optimal AGC command value by constraining the unit command value. The calculation results show that the error of the prediction results of the model proposed in this paper is 5% lower than that of the traditional neural network. The difference between the output value of the optimal AGC command and the expected output value obtained is less than 0.5 MV, which can support AGC online decision-making.© 2017 Elsevier Inc. All rights reserved.

Towards automatic question generation using pre-trained model in academic field for Bahasa Indonesia

AbstractExam evaluations are essential to assessing students’ knowledge and progress in a subject or course. To meet learning objectives and assess student performance, questions must be themed. Automatic Question Generation (AQG) is our novel approach to this problem. A comprehensive process for autonomously generating Bahasa Indonesia text questions is shown. This paper suggests using a decoder to generate text from deep learning models’ tokens. The suggested technique pre-processes Vectorized Corpus, Token IDs, and Features Tensor. The tensors are embedded to increase each token, and attention is masked to separate padding tokens from context-containing tokens. An encoder processes the encoded tokens and attention masks to create a contextual understanding memory that the decoder uses to generate text. Our work uses the Sequence-to-Sequence Learning architecture of BiGRU, BiLSTM, Transformer, BERT, BART, and GPT. Implementing these models optimizes computational resources while extensively exploring the research issue. The model uses context sentences as input and question sentences as output, incorporating linguistic elements like response placement, POS tags, answer masking, and named entities (NE) to improve comprehension and linguistic ability. Our approach includes two innovative models: IndoBERTFormer, which combines a BERT encoder with a Transformer decoder, and IndoBARTFormer, which decodes vectors like BERT. IndoTransGPT uses the Transformer as an encoder to improve understanding, extending the GPT model’s adaptability.

Bloom’s Taxonomy based automatic Marathi question generation

Bloom’s Taxonomy based automatic Marathi question generation

A Survey on Cage‐based Deformation of 3D Models

AbstractInteractive deformation via control handles is essential in computer graphics for the modeling of 3D geometry. Deformation control structures include lattices for free‐form deformation and skeletons for character articulation, but this report focuses on cage‐based deformation. Cages for deformation control are coarse polygonal meshes that encase the to‐be‐deformed geometry, enabling high‐resolution deformation. Cage‐based deformation enables users to quickly manipulate 3D geometry by deforming the cage. Due to their utility, cage‐based deformation techniques increasingly appear in many geometry modeling applications. For this reason, the computer graphics community has invested a great deal of effort in the past decade and beyond into improving automatic cage generation and cage‐based deformation. Recent advances have significantly extended the practical capabilities of cage‐based deformation methods. As a result, there is a large body of research on cage‐based deformation. In this report, we provide a comprehensive overview of the current state of the art in cage‐based deformation of 3D geometry. We discuss current methods in terms of deformation quality, practicality, and precomputation demands. In addition, we highlight potential future research directions that overcome current issues and extend the set of practical applications. In conjunction with this survey, we publish an application to unify the most relevant deformation methods. Our report is intended for computer graphics researchers, developers of interactive geometry modeling applications, and 3D modeling and character animation artists.

Automatic generation method for long-focal-length unobscured freeform optical systems with small volume

The existing design methods for long-focal-length unobscured freeform systems rarely consider the imaging quality requirements and volume constraints simultaneously, causing most of the final designs to not fulfill the requirement of light weight. This study proposes a method to automatically design a long-focal-length unobscured reflective system that satisfies volume constraints while maintaining high imaging quality. First, a method to adaptively set the structural parameter range is proposed, and multiple parameters for different systemic specifications can be effectively calculated within it. Subsequently, the systemic volume and area functions are constructed using the ray tracing method, where the tilt angles, distances between mirrors, and radii of curvature of the mirrors are chosen as the optimization parameters. Third, a comprehensive objective function is jointly established combining ray obscuration and convergence as performance evaluation factors. Then, the structural parameters of a long-focal-length unobscured system with small volume are easily obtained via the simulated annealing method. Finally, the improved W-W method is used to further enhance the imaging quality of the system, and an unobscured freeform reflective optical system with three mirrors is automatically generated. Experimental results demonstrate that our method can automatically calculate the parameter ranges to facilitate the search for structural parameters, and effectively design the long-focal-length unobscured freeform systems with small volume and high imaging quality.

Automatic Generation Technology of Scrap Denim Clothes in Quilting Art Patterns by Integrating Computer Vision.

The combination of computer vision and advanced Fuzzy C-Means algorithms, such as Hybrid OK-Means Fuzzy C-Means (HOFCM), provides a revolutionary way to automatically generate technology of Scrap Denim Clothes in Quilting Art Patterns. The research focuses on the creation of an automatic generating technique specifically designed for making quilting designs from scrap denim clothing. This study investigates the effectiveness of HOFCM in improving the process of creating quilted art patterns from discarded denim materials. Extensive experimentation and analysis have shown that HOFCM dramatically decreases computational time while improving clustering solution quality, particularly when dealing with huge, high-dimensional datasets like scrap denim cloth analysis. The study emphasizes HOFCM's potential to speed up the quilting art pattern generation process, providing designers and artists with a more efficient and precise way to transform old denim clothing into visually appealing designs. The findings highlight HOFCM's transformative impact on automatic generation technology for quilting art patterns, paving the path for innovation and originality in textile art and design. HOFCM significantly lowers time by up to 93.94% compared to regular Fuzzy C-Means on four real and one synthetic datasets. The worst-case scenario resulted in a 2.51% reduction in solution quality.

Image Caption Generator using Deep Learning

Image Caption Generation has always been a study of great interest to the researchers in the Artificial Intelligence department. Being able to program a machine to accurately describe an image or an environment like an average human has major applications in the field of robotic vision, business and many more. Automatic caption generation with attention mechanisms aims at generating more descriptive captions containing coarse to fine semantic contents in the image. This has been a challenging task in the field of artificial intelligence. In this paper, we present different image caption generating models based on deep neural networks, focusing on the various CNN techniques and analyzing their influence on the sentence generation. We have also generated captions for sample images and compared the different feature extraction and encoder models to analyse which model gives better accuracy and generates the desired results

UniSparse: An Intermediate Language for General Sparse Format Customization

The ongoing trend of hardware specialization has led to a growing use of custom data formats when processing sparse workloads, which are typically memory-bound. These formats facilitate optimized software/hardware implementations by utilizing sparsity pattern- or target-aware data structures and layouts to enhance memory access latency and bandwidth utilization. However, existing sparse tensor programming models and compilers offer little or no support for productively customizing the sparse formats. Additionally, because these frameworks represent formats using a limited set of per-dimension attributes, they lack the flexibility to accommodate numerous new variations of custom sparse data structures and layouts. To overcome this deficiency, we propose UniSparse, an intermediate language that provides a unified abstraction for representing and customizing sparse formats. Unlike the existing attribute-based frameworks, UniSparse decouples the logical representation of the sparse tensor (i.e., the data structure) from its low-level memory layout, enabling the customization of both. As a result, a rich set of format customizations can be succinctly expressed in a small set of well-defined query, mutation, and layout primitives. We also develop a compiler leveraging the MLIR infrastructure, which supports adaptive customization of formats, and automatic code generation of format conversion and compute operations for heterogeneous architectures. We demonstrate the efficacy of our approach through experiments running commonly-used sparse linear algebra operations with specialized formats on multiple different hardware targets, including an Intel CPU, an NVIDIA GPU, an AMD Xilinx FPGA, and a simulated processing-in-memory (PIM) device.

Renewable sources with virtual inertia penetration in multi area interconnected AGC using a modified MPC controller

The challenge of controlling frequency deviation becomes more difficult as the complexity of a power plant increases. The robustness of the controller has a major impact on the stability of a complex power system (CPS). Due to the hybridization of CPS basic Automatic Generation Control (AGC) controllers (PID, FOPID, and TID), they are insufficient to give optimal performance to a plant. This requires a robust controller. So, a modified MPC controller has been proposed and evaluated by comparing it with several existing controllers, which gives optimal performance in terms of overshoot, undershoot, and settling time and improves its performance approximately 45%. This research discusses a combined AGC model and control for a three-area CPS, where each producing area consists of a thermal plant, a Diesel power plant, and a Solar Thermal Power Plant (STPP). The results of the modified MPC are superior to those of the basic controller compared to several existing controllers. An improved version of Sea-horse Optimization (SHO) has been proposed to optimize the different controller settings. The superiority of the SHO is shown by a comparison with some well-known, current meta-heuristic methods. The higher penetration levels of renewable energy sources (RESs) reduced system inertia which further deteriorate frequency response in CPS. To overcome these challenges, virtual inertia (VI) has been implemented with MPC. VI is applied to improve the performance of the AGC of the interconnected CPS along with emphasizing the nature of intermittent RESs of PV and wind energy. The thorough study findings provide compelling evidence for the effectiveness and efficiency of the recommended control strategies and also point to the possibility of applying them in actual power systems to improve stability and performance.

Synergizing Deep Learning-Enabled Preprocessing and Human–AI Integration for Efficient Automatic Ground Truth Generation

The progress of incorporating deep learning in the field of medical image interpretation has been greatly hindered due to the tremendous cost and time associated with generating ground truth for supervised machine learning, alongside concerns about the inconsistent quality of images acquired. Active learning offers a potential solution to these problems of expanding dataset ground truth by algorithmically choosing the most informative samples for ground truth labeling. Still, this effort incurs the costs of human labeling, which needs minimization. Furthermore, automatic labeling approaches employing active learning often exhibit overfitting tendencies while selecting samples closely aligned with the training set distribution and excluding out-of-distribution samples, which could potentially improve the model’s effectiveness. We propose that the majority of out-of-distribution instances can be attributed to inconsistent cross images. Since the FDA approved the first whole-slide image system for medical diagnosis in 2017, whole-slide images have provided enriched critical information to advance the field of automated histopathology. Here, we exemplify the benefits of a novel deep learning strategy that utilizes high-resolution whole-slide microscopic images. We quantitatively assess and visually highlight the inconsistencies within the whole-slide image dataset employed in this study. Accordingly, we introduce a deep learning-based preprocessing algorithm designed to normalize unknown samples to the training set distribution, effectively mitigating the overfitting issue. Consequently, our approach significantly increases the amount of automatic region-of-interest ground truth labeling on high-resolution whole-slide images using active deep learning. We accept 92% of the automatic labels generated for our unlabeled data cohort, expanding the labeled dataset by 845%. Additionally, we demonstrate expert time savings of 96% relative to manual expert ground-truth labeling.

Automatic Image Caption Generation with Deep Learning

Artificial intelligence research has long focused on the task of automatically defining what is included in a picture or image. This paper presents the CNN and RNN-LSTM models used in the Automatic Caption Generator implementation. It takes into account recent developments in image processing and performing research on translation through automation. The dataset utilized was Flickr8k. We have employed BLEU scores to assess the efficiency of the system that has been described. Based on the scores, one could classify the resulting captions as excellent or awful. The primary uses of this paradigm are virtual assistants, picture indexing, social media, assistive technology for the blind, altering app suggestions, and many more areas. Key Words: Automatic Caption Generator, CNN, RNN-LSTM, Computer vision, Machine translation, Flickr8k dataset, BLEU scores, Performance evaluation, Virtual assistants, Picture indexing, Social media, Assistive technology for the blind, App suggestions.