IP Geolocation through Reverse DNS

IP Geolocation databases are widely used in online services to map end-user IP addresses to their geographical location. However, they use proprietary geolocation methods, and in some cases they have poor accuracy. We propose a systematic approach to use reverse DNS hostnames for geolocating IP addresses, with a focus on end-user IP addresses as opposed to router IPs. Our method is designed to be combined with other geolocation data sources. We cast the task as a machine learning problem where, for a given hostname, we first generate a list of potential location candidates, and then we classify each hostname and candidate pair using a binary classifier to determine which location candidates are plausible. Finally, we rank the remaining candidates by confidence (class probability) and break ties by population count. We evaluate our approach against three state-of-the-art academic baselines and two state-of-the-art commercial IP geolocation databases. We show that our work significantly outperforms the academic baselines and is complementary and competitive with commercial databases. To aid reproducibility, we open source our entire approach and make it available to the academic community.

Multi-UAV Conflict Resolution with Graph Convolutional Reinforcement Learning

Safety is the primary concern when it comes to air traffic. In-flight safety between Unmanned Aircraft Vehicles (UAVs) is ensured through pairwise separation minima, utilizing conflict detection and resolution methods. Existing methods mainly deal with pairwise conflicts, however, due to an expected increase in traffic density, encounters with more than two UAVs are likely to happen. In this paper, we model multi-UAV conflict resolution as a multiagent reinforcement learning problem. We implement an algorithm based on graph neural networks where cooperative agents can communicate to jointly generate resolution maneuvers. The model is evaluated in scenarios with 3 and 4 present agents. Results show that agents are able to successfully solve the multi-UAV conflicts through a cooperative strategy.

Provable Advantage in Quantum Phase Learning via Quantum Kernel Alphatron

The use of quantum computation to speed-up machine learning algorithms is among the most exciting prospective applications in the NISQ era. Here, we focus on the quantum phase learning problem, which is crucially important in understanding many-particle quantum systems. We prove that, under widely believed complexity theory assumptions, quantum phase learning problem cannot be efficiently solved by machine learning algorithms using classical resources and classical data. Whereas using quantum data, we prove the universality of quantum kernel Alphatron in efficiently predicting quantum phases, indicating clear quantum advantages in such learning problems. We numerically benchmark the algorithm for a variety of problems, including recognizing symmetry-protected topological phases and symmetry-broken phases. Our results highlight the capability of quantum machine learning in efficient prediction of quantum phases of many-particle systems.

The Jigsaw Design Challenge

This article highlights an innovative take on the jigsaw format, an inclusive and cooperative active learning strategy, implemented in an upper-level engineering elective course. After students complete the usual two steps of the jigsaw method—first gaining mastery in “expert groups” and then collaboratively teaching their peers in “jigsaw groups”—they then complete a third step in their jigsaw groups, in which they work together on an authentic design problem, offering a practical take on applying course content. This activity was implemented in three courses offered both in person and remotely (online only). We share how this innovation can promote learning, problem-solving, perspective sharing, and teamwork in contexts with students from different backgrounds and levels of experience.

Configurable Environments in Reinforcement Learning: An Overview

Reinforcement Learning (RL) has emerged as an effective approach to address a variety of complex control tasks. In a typical RL problem, an agent interacts with the environment by perceiving observations and performing actions, with the ultimate goal of maximizing the cumulative reward. In the traditional formulation, the environment is assumed to be a fixed entity that cannot be externally controlled. However, there exist several real-world scenarios in which the environment offers the opportunity to configure some of its parameters, with diverse effects on the agent’s learning process. In this contribution, we provide an overview of the main aspects of environment configurability. We start by introducing the formalism of the Configurable Markov Decision Processes (Conf-MDPs) and we illustrate the solutions concepts. Then, we revise the algorithms for solving the learning problem in Conf-MDPs. Finally, we present two applications of Conf-MDPs: policy space identification and control frequency adaptation.

Chinese Comma Disambiguation in Math Word Problems Using SMOTE and Random Forests

Natural language understanding technologies play an essential role in automatically solving math word problems. In the process of machine understanding Chinese math word problems, comma disambiguation, which is associated with a class imbalance binary learning problem, is addressed as a valuable instrument to transform the problem statement of math word problems into structured representation. Aiming to resolve this problem, we employed the synthetic minority oversampling technique (SMOTE) and random forests to comma classification after their hyperparameters were jointly optimized. We propose a strict measure to evaluate the performance of deployed comma classification models on comma disambiguation in math word problems. To verify the effectiveness of random forest classifiers with SMOTE on comma disambiguation, we conducted two-stage experiments on two datasets with a collection of evaluation measures. Experimental results showed that random forest classifiers were significantly superior to baseline methods in Chinese comma disambiguation. The SMOTE algorithm with optimized hyperparameter settings based on the categorical distribution of different datasets is preferable, instead of with its default values. For practitioners, we suggest that hyperparameters of a classification models be optimized again after parameter settings of SMOTE have been changed.

End-to-End Autonomous Exploration with Deep Reinforcement Learning and Intrinsic Motivation

Developing artificial intelligence (AI) agents is challenging for efficient exploration in visually rich and complex environments. In this study, we formulate the exploration question as a reinforcement learning problem and rely on intrinsic motivation to guide exploration behavior. Such intrinsic motivation is driven by curiosity and is calculated based on episode memory. To distribute the intrinsic motivation, we use a count-based method and temporal distance to generate it synchronously. We tested our approach in 3D maze-like environments and validated its performance in exploration tasks through extensive experiments. The experimental results show that our agent can learn exploration ability from raw sensory input and accomplish autonomous exploration across different mazes. In addition, the learned policy is not biased by stochastic objects. We also analyze the effects of different training methods and driving forces on exploration policy.

PENGGUNAAN METODE RESITASI UNTUK MENINGKATKAN HASIL BELAJAR IPA SISWA KELAS IVA SD NEGERI 192 PEKANBARU

Classroom Action Research (CAR) was carried out to overcome the low science learning outcomes of Class IVa students at SD Negeri 192 Pekanbaru. This science learning problem is solved by applying the recitation method or giving assignment. This Classroom Action Research aims to improve the activities and learning outcomes of Science Class IVa students at SD Negeri 192 Pekanbaru by using the Recitation Method. This classroom action research is expected to be useful for writers, students, schools, and the Pekanbaru City Education Office. Based on research conducted by the author that by using the Resit Method, the learning outcomes of science on the human skeleton, its functions, and maintenance of Grade IVa students of SD Negeri 192 Pekanbaru increased significantly. Prior to the study the classical average was 67.40 or sufficient; in the first cycle to 68.40 (enough); and the result of the second cycle is 79.60 (competent). Completed or achieved KKM individually and classically increased; initially only 16 or 43.20% of students have finished studying; cycle I to 23 students or 62.20%; and in the second cycle as many as 34 students or 91.90%. In cycle II, learning was considered successful because students who achieved the KKM (70) were above 85%. Participants who fail will be given remedial learning. The results of observations, Class IVa students of SD Negeri 192 Pekanbaru using the Assignment Method, students study more diligently and study science more diligently. Based on the results of the study, the Assignment Giving Method succeeded in fixing the problem of low learning outcomes for IVa students at SDN 192 Pekanbaru for the 2019/2020. Penelitian Tindakan Kelas (PTK) ini dilaksanakan untuk mengatasi rendahnya hasil belajar IPA Siswa Kelas IVa SD Negeri 192 Pekanbaru. Masalah belajar IPA ini diatasi dengan menerapkan Metode resitasi atau Pemberian Tugas. Penelitian Tindakan Kelas ini bertujuan untuk Untuk meningkatkan aktivitas dan hasil belajar IPA Siswa Kelas IVa SD Negeri 192 Pekanbaru dengan cara menggunakan Metode Resitasi. Penelitian tindakan Kelas ini diharapkan bermanfaat bagi penulis, siswa, sekolah, dan Dinas Pendidikan Kota Pekanbaru. Berdasarkan penelitian yang dilakukan oleh penulis bahwa dengan menggunakan Metode Resit hasil belajar IPA materi rangka manusia, fungsi, dan pemeliharaannya Siswa Kelas IVa SD Negeri 192 Pekanbaru meningkat secara signifikan. Sebelum penelitian rata-rata secara klasikal adalah 67,40 atau cukup; pada siklus I menjadi 68,40 (cukup); dan hasil siklus II adalah 79,60 (kompeten). Tuntas atau mencapai KKM secara individual dan klasikal meningkat; awalnya hanya 16 atau 43,20% siswa yang tuntas belajar; siklus I menjadi 23 siswa atau 62,20%; dan pada siklus II sebanyak 34 siswa atau 91,90%. Pada siklus II, pembelajaran telah dianggap berhasil karena siswa yang mencapai KKM (70) telah di atas 85%. Peserta yang gagal akan dilakukan pembelajaran remedial. Hasil pengamatan, Siswa Kelas IVa SD Negeri 192 Pekanbaru dengan menggunakan Metode Pemberian Tugas siswa belajar lebih rajin dan tekun belajar IPA. Berdasarkan hasil penelitian pembelajaran Metode Pemberian Tugas berhasil memperbaiki masalah rendahnya hasil belajar siswa IVa SDN 192 Pekanbaru tahun ajaran 2019/2020.