Iterative Compilation Optimization Based on Metric Learning and Collaborative Filtering

Pass selection and phase ordering are two critical compiler auto-tuning problems. Traditional heuristic methods cannot effectively address these NP-hard problems especially given the increasing number of compiler passes and diverse hardware architectures. Recent research efforts have attempted to address these problems through machine learning. However, the large search space of candidate pass sequences, the large numbers of redundant and irrelevant features, and the lack of training program instances make it difficult to learn models well. Several methods have tried to use expert knowledge to simplify the problems, such as using only the compiler passes or subsequences in the standard levels (e.g., -O1, -O2, and -O3) provided by compiler designers. However, these methods ignore other useful compiler passes that are not contained in the standard levels. Principal component analysis (PCA) and exploratory factor analysis (EFA) have been utilized to reduce the redundancy of feature data. However, these unsupervised methods retain all the information irrelevant to the performance of compilation optimization, which may mislead the subsequent model learning. To solve these problems, we propose a compiler pass selection and phase ordering approach, called Iterative Compilation based on Metric learning and Collaborative filtering (ICMC) . First, we propose a data-driven method to construct pass subsequences according to the observed collaborative interactions and dependency among passes on a given program set. Therefore, we can make use of all available compiler passes and prune the search space. Then, a supervised metric learning method is utilized to retain useful feature information for compilation optimization while removing both the irrelevant and the redundant information. Based on the learned similarity metric, a neighborhood-based collaborative filtering method is employed to iteratively recommend a few superior compiler passes for each target program. Last, an iterative data enhancement method is designed to alleviate the problem of lacking training program instances and to enhance the performance of iterative pass recommendations. The experimental results using the LLVM compiler on all 32 cBench programs show the following: (1) ICMC significantly outperforms several state-of-the-art compiler phase ordering methods, (2) it performs the same or better than the standard level -O3 on all the test programs, and (3) it can reach an average performance speedup of 1.20 (up to 1.46) compared with the standard level -O3.

Multimodal Spatio-Temporal Prediction with Stochastic Adversarial Networks

Spatio-temporal (ST) data is a collection of multiple time series data with different spatial locations and is inherently stochastic and unpredictable. An accurate prediction over such data is an important building block for several urban applications, such as taxi demand prediction, traffic flow prediction, and so on. Existing deep learning based approaches assume that outcome is deterministic and there is only one plausible future; therefore, cannot capture the multimodal nature of future contents and dynamics. In addition, existing approaches learn spatial and temporal data separately as they assume weak correlation between them. To handle these issues, in this article, we propose a stochastic spatio-temporal generative model (named D-GAN) which adopts Generative Adversarial Networks (GANs)-based structure for more accurate ST prediction in multiple time steps. D-GAN consists of two components: (1) spatio-temporal correlation network which models spatio-temporal joint distribution of pixels and supports a stochastic sampling of latent variables for multiple plausible futures; (2) a stochastic adversarial network to jointly learn generation and variational inference of data through implicit distribution modeling. D-GAN also supports fusion of external factors through explicit objective to improve the model learning. Extensive experiments performed on two real-world datasets show that D-GAN achieves significant improvements and outperforms baseline models.

Understanding Dilated Mathematical Relationship between Image Features and the Convolutional Neural Network’s Learnt Parameters

Deep learning, in general, was built on input data transformation and presentation, model training with parameter tuning, and recognition of new observations using the trained model. However, this came with a high computation cost due to the extensive input database and the length of time required in training. Despite the model learning its parameters from the transformed input data, no direct research has been conducted to investigate the mathematical relationship between the transformed information (i.e., features, excitation) and the model’s learnt parameters (i.e., weights). This research aims to explore a mathematical relationship between the input excitations and the weights of a trained convolutional neural network. The objective is to investigate three aspects of this assumed feature-weight relationship: (1) the mathematical relationship between the training input images’ features and the model’s learnt parameters, (2) the mathematical relationship between the images’ features of a separate test dataset and a trained model’s learnt parameters, and (3) the mathematical relationship between the difference of training and testing images’ features and the model’s learnt parameters with a separate test dataset. The paper empirically demonstrated the existence of this mathematical relationship between the test image features and the model’s learnt weights by the ANOVA analysis.

PENGGUNAAN METODE JIGSAW UNTUK MENINGKATKAN PRESTASI BELAJAR SISWA PADA MATERI MEDAN MAGNET KELAS IX SMPN 10 MALANG

The purpose of this study was to determine the impact and implementation of the jigsaw method in learning science (Physics) subjects to improve student achievement in class IX of SMP Negeri 10 Malang, Malang City. The method used in this research is classroom action research (CAR). While the data collection techniques used were observation, interviews, documentation and descriptive data analysis. This research was conducted in 3 cycles, namely cycle I, cycle II and cycle III. The research subjects were 34 grade IX students of SMP Negeri 10 Malang in the 2016/2017 academic year. After doing research in class IX SMP Negeri 10 Malang, student learning outcomes using the jigsaw model learning method, can improve student achievement in class IX SMP Negeri 10 Malang. As proof that, from 34 students who carried out learning activities, it was found that the scores of 80, 90 and 100 had increased. Those who scored 80/good enough, in the first cycle: 8 (23.53%) students in the second cycle: 9 (26.47%) students in the third cycle: 13 (38.24%) students. Those who scored 90/good in the first cycle: 1 (2.94%) students in the second cycle: 5 (14.71%) students in the third cycle: 10 (29.41%) students. Those who scored 100/excellent in the first cycle: 0 (0%) students in the second cycle: 0 (0%) students in the third cycle: 1 (2,94%) students. Thus, the jigsaw method of learning can improve the learning outcomes of class IX students in the even semester of 2016/2017 academic year. ABSTRAKTujuan penelitian ini adalah untuk mengetahui dampak dan implementasi metode jigsaw dalam belajar mata pelajaran IPA ( Fisika ) dapat meningkatkan prestasi belajar siswa kelas IX SMP Negeri 10 Malang Kota Malang . Metode yang digunakan dalam penelitian ini adalah metode penelitian tindakan kelas (PTK). Sedangkan teknik pengumpulan data yang digunakan adalah observasi, wawancara, dokumentasi dan analisis data secara deskriptif. Penelitian ini dilakukan dalam 3 siklus yakni siklus I siklus II dan siklus III. Subjek penelitian adalah 34 orang siswa kelas IX SMP Negeri 10 Malang Tahun Pelajaran 2016/2017. Setelah dilakukan penelitian di kelas IX SMP Negeri 10 Malang, hasil belajar siswa dengan menggunakan metode pembelajaran Model jigsaw, dapat meningkatkan prestasi belajar Siswa Kelas IX SMP Negeri 10 Malang. Sebagai buktinya bahwa, dari 34 siswa yang melakukan kegiatan belajar didapatkan pada perolehan nilai 80 , 90 dan 100 mengalami peningkatan. Yang memperoleh nilai 80/cukup baik, pada siklus I : 8 (23,53%) siswa pada siklus II : 9 (26,47%) siswa pada siklus III : 13 (38,24%) siswa. Yang memperoleh nilai 90/baik pada siklus I : 1 (2,94%) siswa pada siklus II : 5 (14,71%) siswa pada siklus III : 10 (29,41%) siswa. Yang memperoleh nilai 100/sangat baik pada siklus I : 0 (0%) siswa pada siklus II : 0 (0%) siswa pada siklus III : 1 (2,94%) siswa. Dengan demikian, metode pembelajaran dengan metode jigsaw dapat meningkatkan hasil belajar siswa kelas IX semester genap tahun pelajaran 2016/2017.

Development of Manakarra Value Integrative Model in Learning Introduction to Education Course

This study aims to produce a valid, practical, and effective Manakarra value integrative model learning tool that students can implement in the Introduction to Education course. The development of learning tools using the 4-D model and tested on students of the Faculty of Teacher Training and Education, Tomakaka University, Study Program of Indonesian Language and Literature Education. The research design used One-Group Pretest-Posttest Design. Data collection was using observation, tests, and questionnaires. The data analysis technique used quantitative/qualitative descriptive analysis. Learning the Manakarra value integrative model begins with the validity of learning tools; Model Books, Teaching Materials, Semester Learning Plans (RPS), Student Activity Sheets and Learning Outcomes Test Instruments. Furthermore, a trial was conducted on students in learning the Introduction to Education course. The implementation of the learning stages seen in student learning activities showed a very good improvement and enthusiasm in carrying out learning. Students can follow the learning process by using learning tools developed with high curiosity. The effectiveness of the Manakarra value integrative model from the trial results shows that the integrative model's learning process has succeeded in improving students' critical thinking skills.

Model Learning Approach pada Mata Kuliah Basic Aircraft Material (BAM) di Perguruan Tinggi Vokasi Indonesia

Mata kuliah Basic Aircraft Material (BAM) adalah bidang ilmu yang mempelajari tentang dasar-dasar pemeliharaan pesawat udara. Sistem penilaian yang dilakukan pada mata kuliah ini selain dari UTS dan UAS, juga berdasarkan 2 tugas yang dikerjakan, dan sikap Taruna selama mempelajari mata kuliah BAM. Sedangkan tuntutan pada mata kuliah BAM ini adalah Taruna dituntut untuk dapat menganalisis kasus serta melakukan praktek kerja. Agar Taruna dapat memenuhi tuntutan mata kuliah BAM tersebut, maka Taruna memiliki pendekatan yang berbeda-beda dalam belajar. Hasil penelitian ditemukan bahwa penbelajaran taruna dengan pendekatan belajar surface approach, sebesar 40% taruna menyatakan mata kuliah tersebut penting dan 60% menyatakan bahwa tugas yang diberikan merupakan beban dan hanya belajar denganberfokus pada sub pokok bahasan yang dianggap penting saja. Kesimpulan dengan metode pembelajaran ini, maka hasil pembelajaran taruna semakin meningkat.

On the Neural Network Solution of One-Dimensional Wave Problem

Artificial neural network has become an emerging popular method to handle various problems, especially in case where it has deep multiple neural layers. In this study, we use a deep artificial neural network model to solve one-dimensional wave equation, without any external datasets. Different type of boundary conditions, i.e., Dirichlet, Neumann, and Robin, are used. We analyze the model learning capabilities in a set of settings, such as data setup and the model width and depth. We also present some discussions of advantages and disadvantages of the model in comparison with other matured existing techniques to solve wave equation.

Generative Adversarial Networks for Morphological–Temporal Classification of Stem Cell Images

Frequently, neural network training involving biological images suffers from a lack of data, resulting in inefficient network learning. This issue stems from limitations in terms of time, resources, and difficulty in cellular experimentation and data collection. For example, when performing experimental analysis, it may be necessary for the researcher to use most of their data for testing, as opposed to model training. Therefore, the goal of this paper is to perform dataset augmentation using generative adversarial networks (GAN) to increase the classification accuracy of deep convolutional neural networks (CNN) trained on induced pluripotent stem cell microscopy images. The main challenges are: 1. modeling complex data using GAN and 2. training neural networks on augmented datasets that contain generated data. To address these challenges, a temporally constrained, hierarchical classification scheme that exploits domain knowledge is employed for model learning. First, image patches of cell colonies from gray-scale microscopy images are generated using GAN, and then these images are added to the real dataset and used to address class imbalances at multiple stages of training. Overall, a 2% increase in both true positive rate and F1-score is observed using this method as compared to a straightforward, imbalanced classification network, with some greater improvements on a classwise basis. This work demonstrates that synergistic model design involving domain knowledge is key for biological image analysis and improves model learning in high-throughput scenarios.