Naive Approach Research Articles

The bubble of nothing under T-duality

The bubble of nothing is a solution to Einstein’s equations where a circle shrinks and pinches off smoothly. As such, it is one of the simplest examples of a dynamical cobordism to nothing. We take a first step in studying how this solution transforms under T-duality in bosonic string theory. Applying Buscher’s rules reveals that the dual solution features a singular, strongly coupled core, with a circle blowing-up rather than pinching off. This naive approach to T-duality solely accounts for the zero-modes of the fields after dimensional reduction on the circle. For this reason, we argue that this is not the full picture that the T-dual solution should depend non-trivially on the dual circle. We point out evidence to this effect both in the gravity description and on the worldsheet. A more complete description of the T-dual object would require a full-fledged sigma model for the bubble of nothing. Nevertheless, inspired by similar examples in the literature, we detail one possible scenario where the stringy bubble of nothing is mediated by closed string tachyon condensation and we discuss its T-duality.

IAQ Prediction in Apartments Using Machine Learning Techniques and Sensor Data

This study explores the capability of machine learning techniques (MLTs) in predicting IAQ in apartments. Sensor data from kitchen air monitoring were used to determine the conditions in the living room. The analysis was based on several air parameters—temperature, relative humidity, CO2 concentration, and TVOC—recorded in five apartments. Multiple input–multiple output prediction models were built. Linear (multiple linear regression and multilayer perceptron (MLP)) and nonlinear (decision trees, random forest, k-nearest neighbors, and MLP) methods were investigated. Five-fold cross-validation was applied, where four apartments provided data for model training and the remaining one was the source of the test data. The models were compared using performance metrics (R2, MAPE, and RMSE). The naive approach was used as the benchmark. This study showed that linear MLTs performed best. In this case, the coefficients of determination were highest: R2 = 0.94 (T), R2 = 0.94 (RH), R2 = 0.63 (CO2), R2 = 0.84 (TVOC, based on the SGP30 sensor), and R2 = 0.92 (TVOC, based on the SGP30 sensor). The prediction of distinct indoor air parameters was not equally effective. Based on the lowest percentage error, best predictions were attained for indoor air temperature (MAPE = 1.57%), relative humidity (MAPE = 2.97%RH), and TVOC content (MAPE = 0.41%). Unfortunately, CO2 prediction was loaded with high error (MAPE = 20.83%). The approach was particularly effective in open-kitchen apartments, and they could be the target for its application. This research offers a method that could contribute to attaining effective IAQ control in apartments.

Benchmarking reservoir computing for residential energy demand forecasting

In the energy sector, accurate demand forecasts are vital but often limited by the available computational power. Reservoir computing (RC) or echo-state networks excel in chaotic time series prediction, with lower computational requirements compared to other recurrent network based methods like LSTMs. Next-generation reservoir computing (NG-RC) is a newer, more efficient variant of classical RC originating from nonlinear vector autoregression and therefore missing the randomness of classical RC. In our study, we evaluate RC and NG-RC for day-ahead energy demand predictions on four data sets and compare it to LSTMs and a naive persistence approach. We find that NG-RC outperforms all other methods when considering the root mean squared error on all data sets but struggles with very small or zero demands. Additionally, it offers a very computationally effective hyperparameter optimization and excels in replicating the inherent volatility and the erratic behavior of energy demands.

BALANCING SARCASTIC HINGLISH SHORT TEXT DATA USING AUGMENTATION TECHNIQUES WITH HANDLING SPELLING VARIATIONS

In the real world, there is a significant presence of imbalanced data due to the fact that the classes that make up the datasets are not evenly distributed. Even when using methods that are traditionally used to achieve class balance, such as re-sampling & re-weighting, current deep learning still faces a significant obstacle because of the class imbalance. This study’s major objective is proposing a data augmentation technique to balance the data to improve the sample sizes for the minority classes. Python, a well-known programming language, & multiple methods of machine learning are being employed in the execution of this study. Classification models like Logistic Regression, Naïve Bayes, Support Vector Machine, Decision Tree, Random Forest, Extra Trees Classifier, AdaBoost classifier, Gradient Boost classifier was used to implement this study. Precision, recall, & F-score were used to determine which model would be the most effective. According to the findings of this study's analysis, the Naive Bayes approach, which has a F1-Score of 95.85% & has Wn = 3, Cn = 3, & CWn =3 as its parameters, is the technique that yields the most accurate results.

Perbandingan Pendekatan Metode Peramalan Naive Approach, Simple Moving Average dan Weighted Moving Average dalam Upaya Meningkatkan Prediksi Penjualan JNE Kopma UNY

JNE Kopma UNY's logistics demand increases 2% per week, the manual courier allocation system is unable to keep up, predictions are accurate and real-time is needed. However, there is no exact accuracy for this case. The study was conducted to determine the accuracy of the smallest Mean Percentage Error (MPE) at JNE Kopma UNY by observing sales data for 2022 to predict demand. The data was analyzed using the Naive Approach, SMA and WMA methods, then measured using the MPE to determine the best forecasting method. The software used is Microsoft Excel 2016. The 3 Month (SMA) produces the highest sales prediction (Rp. 38,668,850) for January 2023 compared to Naive Approach (Rp. 35,086,330) and 3 Month WMA (Rp. 37,993,380). 3 Month SMA is proven to be superior with the lowest MPE (12.07%) and consistent performance. Naive Approach is inaccurate (MPE 20.26%) and 3 Month WMA performance is unstable (MPE 14.27%). This research recommends 3 Month SMA for JNE Kopma UNY to increase the accuracy of sales predictions and optimize stock. Opportunities for further research are open regarding the application of 3 Month SMA in various contexts.

Text Mining for News Forecasting on The Turnback Hoax Website

News has been disseminated swiftly via the internet due to the rapid growth of information technology. The rapid spreading of news often confuses because the truth cannot be ascertained. Additionally, online social media is becoming increasingly popular, making it an excellent environment for propagating false information, including misinformation, phony reviews, advertising, rumors, political remarks, innuendo, etc. This study's specific goal is to classify data using a data mining approach model called text mining so that a system can automatically do the classification. As a result, the study will produce a dataset, which can then be used to create an application using data mining's ability to predict breaking news. An application was produced by employing data mining to forecast recent news. This study was able to classify data using a naive Bayes data mining approach model so that a system can automatically do the classification. The study produced an accuracy of 77% obtained with training data of 82%. From 994 contents, the classification of misleading content reached 33.9%, false content as many as 24.85%, imitation content was 13.48%, fake content reached 11.07%, manipulated content was 9.86%, parody content was 3.22%, satire content was 2.31%, and connection content as many as 1.31%. This study then visualizes the results using bar charts and word clouds. This work also produced datasets with the naïve Bayes method of news data and news that has been valid. Afterward, the dataset will be used in making applications to produce prototypes of computer program applications.

Analysis of Fruit Consumer Satisfaction in an Online Purchasing System Using an E-commerce Platform: A Naive Bayes Approach

E-commerce platforms have now developed into fresh agricultural products, one of which is fruit. Purchasing online means customers cannot choose their own products like in offline stores. Therefore, the aim of this research is to identify the attributes that consumers often talk about and analyze consumer satisfaction with fruit on e-commerce platforms. Data analysis in this research was carried out using descriptive statistical methods and the Naïve Bayes approach. Data was taken from the Tokopedia platform with a total of 316 reviews using random sampling. The research results show that the attributes most frequently reviewed by consumers are service attributes (49%) and product attributes (44%), while the price attribute is reviewed the least (7%). Attributes related to product, price and service were assessed as having negative performance overall because nine of the twelve attributes analyzed indicated consumer dissatisfaction. Seller reviews are very important for business continuity so that sellers can improve their services and pay more attention to product quality so that the reviews appear positive. Research using the naive Bayes approach to determine customer satisfaction sentiment towards fruit products has not been widely carried out so the results of this research can complement previous studies.

Doubly Perturbed Task Free Continual Learning

Task-free online continual learning (TF-CL) is a challenging problem where the model incrementally learns tasks without explicit task information. Although training with entire data from the past, present as well as future is considered as the gold standard, naive approaches in TF-CL with the current samples may be conflicted with learning with samples in the future, leading to catastrophic forgetting and poor plasticity. Thus, a proactive consideration of an unseen future sample in TF-CL becomes imperative. Motivated by this intuition, we propose a novel TF-CL framework considering future samples and show that injecting adversarial perturbations on both input data and decision-making is effective. Then, we propose a novel method named Doubly Perturbed Continual Learning (DPCL) to efficiently implement these input and decision-making perturbations. Specifically, for input perturbation, we propose an approximate perturbation method that injects noise into the input data as well as the feature vector and then interpolates the two perturbed samples. For decision-making process perturbation, we devise multiple stochastic classifiers. We also investigate a memory management scheme and learning rate scheduling reflecting our proposed double perturbations. We demonstrate that our proposed method outperforms the state-of-the-art baseline methods by large margins on various TF-CL benchmarks.

Analysis of Differentially Private Synthetic Data: A Measurement Error Approach

Differentially private (DP) synthetic datasets have been receiving significant attention from academia, industry, and government. However, little is known about how to perform statistical inference using DP synthetic datasets. Naive approaches that do not take into account the induced uncertainty due to the DP mechanism will result in biased estimators and invalid inferences. In this paper, we present a class of maximum likelihood estimator (MLE)-based easy-to-implement bias-corrected DP estimators with valid asymptotic confidence intervals (CI) for parameters in regression settings, by establishing the connection between additive DP mechanisms and measurement error models. Our simulation shows that our estimator has comparable performance to the widely used sufficient statistic perturbation (SSP) algorithm in some scenarios but with the advantage of releasing a synthetic dataset and obtaining statistically valid asymptotic CIs, which can achieve better coverage when compared to the naive CIs obtained by ignoring the DP mechanism.

Class-Attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective

Modern classification problems exhibit heterogeneities across individual classes: Each class may have unique attributes, such as sample size, label quality, or predictability (easy vs difficult), and variable importance at test-time. Without care, these heterogeneities impede the learning process, most notably, when optimizing fairness objectives. Confirming this, under a gaussian mixture setting, we show that the optimal SVM classifier for balanced accuracy needs to be adaptive to the class attributes. This motivates us to propose CAP: An effective and general method that generates a class-specific learning strategy (e.g.~hyperparameter) based on the attributes of that class. This way, optimization process better adapts to heterogeneities. CAP leads to substantial improvements over the naive approach of assigning separate hyperparameters to each class. We instantiate CAP for loss function design and post-hoc logit adjustment, with emphasis on label-imbalanced problems. We show that CAP is competitive with prior art and its flexibility unlocks clear benefits for fairness objectives beyond balanced accuracy. Finally, we evaluate CAP on problems with label noise as well as weighted test objectives to showcase how CAP can jointly adapt to different heterogeneities.

Optimal Decision Strategies for the Generalized Cuckoo Card Game

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Cuckoo</i> is a popular card game, which originated in France during the 15th century and then spread throughout Europe, where it is currently well-known under distinct names and with different variants. Cuckoo is an imperfect information game-of-chance, which makes the research regarding its optimal strategies determination interesting. The rules are simple: each player receives a covered card from the dealer; starting from the player at the dealer's left, each player looks at its own card and decides whether to exchange it with the player to their left, or keep it; the dealer plays at last and, if it decides to exchange card, it draws a random one from the remaining deck; the player(s) with the lowest valued card lose(s) the round. We formulate the gameplay mathematically and provide an analysis of the optimal decision policies. Different card decks can be used for this game, e.g., the standard 52-card deck or the Italian 40-card deck. We generalize the decision model for an arbitrary number decks' cards, suites, and players. Lastly, through numerical simulations, we compare the determined optimal decision strategy against different benchmarks, showing that the strategy outperforms the random and naive policies and approaches the performance of the ideal oracle.

Generic Machine Learning Framework for Fully-Unsupervised Anomaly Detection with Contaminated Data

Anomaly detection (AD) tasks have been solved using machine learning algorithms in various domains and applications. The great majority of these algorithms use normal data to train a residual-based model, and assign anomaly scores to unseen samples based on their dissimilarity with the learned normal regime. The underlying assumption of these approaches is that anomaly-free data is available for training. This is, however, often not the case in real-world operational settings, where the training data may be contaminated with a certain fraction of abnormal samples. Training with contaminated data, in turn, inevitably leads to a deteriorated AD performance of the residual-based algorithms.&#x0D; In this paper we introduce a framework for a fully unsupervised refinement of contaminated training data for AD tasks. The framework is generic and can be applied to any residual-based machine learning model. We demonstrate the application of the framework to two public datasets of multivariate time series machine data from different application fields. We show its clear superiority over the naive approach of training with contaminated data without refinement. Moreover, we compare it to the ideal, unrealistic reference in which anomaly-free data would be available for training. Since the approach exploits information from the anomalies, and not only from the normal regime, it is comparable and often outperforms the ideal baseline as well.

Identifying similar-bicliques in bipartite graphs

Bipartite graphs have been widely used to model the relationship between entities of different types, where vertices are partitioned into two disjoint sets/sides. Finding dense subgraphs in a bipartite graph is of great significance and encompasses many applications. However, none of the existing dense bipartite subgraph models consider similarity between vertices from the same side, and as a result, the identified results may include vertices that are not similar to each other. In this work, we formulate the notion of similar-biclique which is a special kind of biclique where all vertices from a designated side are similar to each other and aim to enumerate all similar-bicliques. The naive approach of first enumerating all maximal bicliques and then extracting all maximal similar-bicliques from them is inefficient, as enumerating maximal bicliques is already time consuming. We propose a backtracking algorithm MSBE\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ extsf{MSBE}$$\\end{document} to directly enumerate maximal similar-bicliques and power it by vertex reduction and optimization techniques. In addition, we design a novel index structure to speed up a time-critical operation of MSBE\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ extsf{MSBE}$$\\end{document}, as well as to speed up vertex reduction. Efficient index construction algorithms are developed. To handle dynamic graph updates, we also propose algorithms and optimization techniques for maintaining our index. Finally, we parallelize our index construction algorithms to exploit multiple CPU cores. Extensive experiments on 17 bipartite graphs as well as case studies are conducted to demonstrate the effectiveness and efficiency of our model and algorithms.

Forward Learning of Large Language Models by Consumer Devices

Large Language Models achieve state of art performances on a broad variety of Natural Language Processing tasks. In the pervasive IoT era, their deployment on edge devices is more compelling than ever. However, their gigantic model footprint has hindered on-device learning applications which enable AI models to continuously learn and adapt to changes over time. Back-propagation, in use by the majority of deep learning frameworks, is computationally intensive and requires storing intermediate activations into memory to cope with the model’s weights update. Recently, “Forward-only algorithms” have been proposed since they are biologically plausible alternatives. By applying more “forward” passes, this class of algorithms can achieve memory reductions with respect to more naive forward-only approaches and by removing the need to store intermediate activations. This comes at the expense of increased computational complexity. This paper considered three Large Language Model: DistilBERT, GPT-3 Small and AlexaTM. It investigated quantitatively any improvements about memory usage and computational complexity brought by known approaches named PEPITA and MEMPEPITA with respect to backpropagation. For low number of tokens in context, and depending on the model, PEPITA increases marginally or reduces substantially arithmetic operations. On the other hand, for large number of tokens in context, PEPITA reduces computational complexity by 30% to 50%. MEMPEPITA increases PEPITA’s complexity by one third. About memory, PEPITA and backpropagation, require a comparable amount of memory to store activations, while MEMPEPITA reduces it by 50% to 94% with the benefits being more evident for architectures with a long sequence of blocks. In various real case scenarios, MEMPEPITA’s memory reduction was essential for meeting the tight memory requirements of 128 MB equipped edge consumer devices, which are commonly available as smartphone and industrial application multi processors.

Mesh Compression Algorithm for Geometrical Coordinates in Computational Meshes

Application of advanced mesh based methods, including adaptive finite element method, is impossible without theoretical elaboration and practical realization of a model for organization and functionality of computational mesh. One of the most basic mesh functionality is storing and providing geometrical coordinates for vertices and other mesh entities. New algorithm for this task based on on-the-fly recreation of coordinates was developed. Conducted tests are proving that, for selected cases, it can be orders of magnitude faster than naive approach or other similar algorithms.

CompTrails: comparing hypotheses across behavioral networks

AbstractThe term Behavioral Networks describes networks that contain relational information on human behavior. This ranges from social networks that contain friendships or cooperations between individuals, to navigational networks that contain geographical or web navigation, and many more. Understanding the forces driving behavior within these networks can be beneficial to improving the underlying network, for example, by generating new hyperlinks on websites, or by proposing new connections and friends on social networks. Previous approaches considered different hypotheses on a single network and evaluated which hypothesis fits best. These hypotheses can represent human intuition and expert opinions or be based on previous insights. In this work, we extend these approaches to enable the comparison of a single hypothesis between multiple networks. We unveil several issues of naive approaches that potentially impact comparisons and lead to undesired results. Based on these findings, we propose a framework with five flexible components that allow addressing specific analysis goals tailored to the application scenario. We show the benefits and limits of our approach by applying it to synthetic data and several real-world datasets, including web navigation, bibliometric navigation, and geographic navigation. Our work supports practitioners and researchers with the aim of understanding similarities and differences in human behavior between environments.

Development of Two Dimension (2D) Game Engine with Finite State Machine (FSM) Based Artificial Intelligence (AI) Subsystem

With Al becoming more and more relevant in today’s world, this project aims to develop a 2D game engine with an Al subsystem for state-driven agents, which is rarely implemented by a lot of 2D engines out there. In this study, a 2D game engine was designed with an FSM (Finite State Machine)--based AL subsystem using state-driven game agents. The engine was implemented using the Javascript programming language and the WebGL 2.0 graphics library/API. It is targeted at web-based games/simulations. Components and subsystems include physics, audio, math, rendering, and AI (based on finite-state machines). The FSM-based AI subsystem is a solution aimed at reducing the ambiguity and performance hits associated with creating 2D game AI in the naive approach. The AI subsystem creates an interface for 2D games to be created with a common paradigm, and simulated with a great level of realism. The state machine used in this study is used to represent a variety of behaviours, such as wandering, attacking, and fleeing. The following conclusions were drawn as regards the impact of the AI approach used on rendering performance; the naive approach to implementing game AI is also compared with the FSM approach in terms of rendering (frames per second/ FPS, or frame rate). With the naive approach (vector math) being used to implement AI, there was a drop in the rendering frame rate to 50 FPS. The FSM approach didn’t affect the frame rate, which is usually at 60FPS. With a well-developed FSM (Finite State Machine), game agents can transition between states easily as a response to user input or stimulus from the game environment. The proposed method was tested by creating a prototype game with the 2D game engine. The prototype game was a straightforward side-scrolling platformer featuring a cast of non-player characters (NPCs). This approach to implementing 2D game AI goes a long way toward improving performance and mitigating ambiguity in the code.

Measurements With A Quantum Vision Transformer: A Naive Approach

In mainstream machine learning, transformers are gaining widespread usage. As Vision Transformers rise in popularity in computer vision, they now aim to tackle a wide variety of machine learning applications. In particular, transformers for High Energy Physics (HEP) experiments continue to be investigated for tasks including jet tagging, particle reconstruction, and pile-up mitigation. An improved Quantum Vision Transformer (QViT) with a quantum-enhanced self-attention mechanism is introduced and discussed. A shallow circuit is proposed for each component of self-attention to leverage current Noisy Intermediate Scale Quantum (NISQ) devices. Variations of the hybrid architecture/model are explored and analyzed. The results demonstrate a successful proof of concept for the QViT, and establish a competitive performance benchmark for the proposed design and implementation. The findings also provide strong motivation to experiment with different architectures, hyperparameters, and datasets, setting the stage for implementation in HEP environments where transformers are increasingly used in state of the art machine learning solutions.

A Practical Contrastive Learning Framework for Single-Image Super-Resolution.

Contrastive learning has achieved remarkable success on various high-level tasks, but there are fewer contrastive learning-based methods proposed for low-level tasks. It is challenging to adopt vanilla contrastive learning technologies proposed for high-level visual tasks to low-level image restoration problems straightly. Because the acquired high-level global visual representations are insufficient for low-level tasks requiring rich texture and context information. In this article, we investigate the contrastive learning-based single-image super-resolution (SISR) from two perspectives: positive and negative sample construction and feature embedding. The existing methods take naive sample construction approaches (e.g., considering the low-quality input as a negative sample and the ground truth as a positive sample) and adopt a prior model (e.g., pretrained very deep convolutional networks proposed by visual geometry group (VGG) model) to obtain the feature embedding. To this end, we propose a practical contrastive learning framework for SISR (PCL-SR). We involve the generation of many informative positive and hard negative samples in frequency space. Instead of utilizing an additional pretrained network, we design a simple but effective embedding network inherited from the discriminator network, which is more task-friendly. Compared with the existing benchmark methods, we retrain them by our proposed PCL-SR framework and achieve superior performance. Extensive experiments have been conducted to show the effectiveness and technical contributions of our proposed PCL-SR thorough ablation studies. The code and resulting models will be released via https://github.com/Aitical/PCL-SISR.

Memory-efficient neurons and synapses for spike-timing-dependent-plasticity in large-scale spiking networks.

This paper addresses the challenges posed by frequent memory access during simulations of large-scale spiking neural networks involving synaptic plasticity. We focus on the memory accesses performed during a common synaptic plasticity rule since this can be a significant factor limiting the efficiency of the simulations. We propose neuron models that are represented by only three state variables, which are engineered to enforce the appropriate neuronal dynamics. Additionally, memory retrieval is executed solely by fetching postsynaptic variables, promoting a contiguous memory storage and leveraging the capabilities of burst mode operations to reduce the overhead associated with each access. Different plasticity rules could be implemented despite the adopted simplifications, each leading to a distinct synaptic weight distribution (i.e., unimodal and bimodal). Moreover, our method requires fewer average memory accesses compared to a naive approach. We argue that the strategy described can speed up memory transactions and reduce latencies while maintaining a small memory footprint.