Search

This paper presents novel game AI agents based on the AlphaZero framework, enhanced with Vision Transformers (ViT): AlphaViT, AlphaViD, and AlphaVDA. These agents are designed to play various board games of different sizes using a single model, overcoming AlphaZero's limitation of being restricted to a fixed board size. AlphaViT uses only a transformer encoder, while AlphaViD and AlphaVDA contain both an encoder and a decoder. AlphaViD's decoder receives input from the encoder output, while AlphaVDA uses a learnable matrix as decoder input. Using the AlphaZero framework, the three proposed methods demonstrate their versatility in different game environments, including Connect4, Gomoku, and Othello. Experimental results show that these agents, whether trained on a single game or on multiple games simultaneously, consistently outperform traditional algorithms such as Minimax and Monte Carlo tree search using a single DNN with shared weights, while approaching the performance of AlphaZero. In particular, AlphaViT and AlphaViD show strong performance across games, with AlphaViD benefiting from an additional decoder layer that enhances its ability to adapt to different action spaces and board sizes. These results may suggest the potential of transformer-based architectures to develop more flexible and robust game AI agents capable of excelling in multiple games and dynamic environments.

This research aims to develop kernel GNG, a kernelized version of the growing neural gas (GNG) algorithm, and to investigate the features of the networks generated by the kernel GNG. The GNG is an unsupervised artificial neural network that can transform a dataset into an undirected graph, thereby extracting the features of the dataset as a graph. The GNG is widely used in vector quantization, clustering, and 3D graphics. Kernel methods are often used to map a dataset to feature space, with support vector machines being the most prominent application. This paper introduces the kernel GNG approach and explores the characteristics of the networks generated by kernel GNG. Five kernels, including Gaussian, Laplacian, Cauchy, inverse multiquadric, and log kernels, are used in this study. The results of this study show that the average degree and the average clustering coefficient decrease as the kernel parameter increases for Gaussian, Laplacian, Cauchy, and IMQ kernels. If we avoid more edges and a higher clustering coefficient (or more triangles), the kernel GNG with a larger value of the parameter will be more appropriate.

The growth of network-connected devices has led to an exponential increase in data generation, creating significant challenges for efficient data analysis. This data is generated continuously, creating a dynamic flow known as a data stream. The characteristics of a data stream may change dynamically, and this change is known as concept drift. Consequently, a method for handling data streams must efficiently reduce their volume while dynamically adapting to these changing characteristics. This paper proposes a simple online vector quantization method for concept drift. The proposed method identifies and replaces units with low win probability through remove-birth updating, thus achieving a rapid adaptation to concept drift. Furthermore, the results of this study show that the proposed method can generate minimal dead units even in the presence of concept drift. This study also suggests that some metrics calculated from the proposed method will be helpful for drift detection.

Mars exploration motivates the search for extraterrestrial life, the development of space technologies, and the design of human missions and habitations. Here we seek new insights and pose unresolved questions relating to the natural history of Mars, habitability, robotic and human exploration, planetary protection, and the impacts on human society. Key observations and findings include:(1)high escape rates of early Mars' atmosphere, including loss of water, impact present-day habitability;(2)putative fossils on Mars will likely be ambiguous biomarkers for life;(3)microbial contamination resulting from human habitation is unavoidable;(4)based on Mars' current planetary protection category, robotic payload(s) should characterize the local martian environment for any life-forms prior to human habitation. Some of the outstanding questions are:(1)which interpretation of the hemispheric dichotomy of the planet is correct;(2)to what degree did deep-penetrating faults transport subsurface liquids to Mars' surface;(3)in what abundance are carbonates formed by atmospheric processes;(4)what properties of martian meteorites could be used to constrain their source locations;(5)the origin(s) of organic macromolecules;(6)was/is Mars inhabited;(7)how can missions designed to uncover microbial activity in the subsurface eliminate potential false positives caused by microbial contaminants from Earth;(8)how can we ensure that humans and microbes form a stable and benign biosphere;(9)should humans relate to putative extraterrestrial life from a biocentric viewpoint (preservation of all biology), or anthropocentric viewpoint of expanding habitation of space? Studies of Mars' evolution can shed light on the habitability of extrasolar planets. In addition, Mars exploration can drive future policy developments and confirm (or put into question) the feasibility and/or extent of human habitability of space.

Artificial intelligence (AI) has achieved superhuman performance in board games such as Go, chess, and Othello (Reversi). In other words, the AI system surpasses the level of a strong human expert player in such games. In this context, it is difficult for a human player to enjoy playing the games with the AI. To keep human players entertained and immersed in a game, the AI is required to dynamically balance its skill with that of the human player. To address this issue, we propose AlphaDDA, an AlphaZero-based AI with dynamic difficulty adjustment (DDA). AlphaDDA consists of a deep neural network (DNN) and a Monte Carlo tree search, as in AlphaZero. AlphaDDA learns and plays a game the same way as AlphaZero, but can change its skills. AlphaDDA estimates the value of the game state from only the board state using the DNN. AlphaDDA changes a parameter dominantly controlling its skills according to the estimated value. Consequently, AlphaDDA adjusts its skills according to a game state. AlphaDDA can adjust its skill using only the state of a game without any prior knowledge regarding an opponent. In this study, AlphaDDA plays Connect4, Othello, and 6x6 Othello with other AI agents. Other AI agents are AlphaZero, Monte Carlo tree search, the minimax algorithm, and a random player. This study shows that AlphaDDA can balance its skill with that of the other AI agents, except for a random player. The DDA ability of AlphaDDA is based on an accurate estimation of the value from the state of a game. We believe that the AlphaDDA approach for DDA can be used for any game AI system if the DNN can accurately estimate the value of the game state and we know a parameter controlling the skills of the AI system., Comment: 24 pages, 10 figures

Spherical data is distributed on the sphere. The data appears in various fields such as meteorology, biology, and natural language processing. However, a method for analysis of spherical data does not develop enough yet. One of the important issues is an estimation of the number of clusters in spherical data. To address the issue, I propose a new method called the Spherical X-means (SX-means) that can estimate the number of clusters on d-dimensional sphere. The SX-means is the model-based method assuming that the data is generated from a mixture of von Mises-Fisher distributions. The present paper explains the proposed method and shows its performance of estimation of the number of clusters.

Spectral clustering (SC) is one of the most popular clustering methods and often outperforms traditional clustering methods. SC uses the eigenvectors of a Laplacian matrix calculated from a similarity matrix of a dataset. SC has serious drawbacks: the significant increases in the time complexity derived from the computation of eigenvectors and the memory space complexity to store the similarity matrix. To address the issues, I develop a new approximate spectral clustering using the network generated by growing neural gas (GNG), called ASC with GNG in this study. ASC with GNG uses not only reference vectors for vector quantization but also the topology of the network for extraction of the topological relationship between data points in a dataset. ASC with GNG calculates the similarity matrix from both the reference vectors and the topology of the network generated by GNG. Using the network generated from a dataset by GNG, ASC with GNG achieves to reduce the computational and space complexities and improve clustering quality. In this study, I demonstrate that ASC with GNG effectively reduces the computational time. Moreover, this study shows that ASC with GNG provides equal to or better clustering performance than SC.

Throughout the history of the solar system, Mars has experienced continuous asteroidal impacts. These impacts have produced impact-generated Mars ejecta, and a fraction of this debris is delivered to Earth as Martian meteorites. Another fraction of the ejecta is delivered to the moons of Mars, Phobos and Deimos. Here, we studied the amount and condition of recent delivery of impact ejecta from Mars to its moons. Using state-of-the-art numerical approaches, we report, for the first time, that materials delivered from Mars to its moons are physically and chemically different from the Martian meteorites, which are all igneous rocks with a limited range of ages. We show that Mars ejecta mixed in the regolith of its moons potentially covers all its geological eras and consists of all types of rocks, from sedimentary to igneous. A Martian moons sample-return mission will bring such materials back to Earth, and the samples will provide a wealth of "time-resolved" geochemical information about the evolution of Martian surface environments., Comment: 9 pages, 3 figures, published in Scientific Reports

This paper presents a case study of microbe transportation in the Mars-satellites system. We examined the spatial distribution of potential impact-transported microbes on the Martian moons using impact physics by following a companion study (Fujita et al.). We used sterilization data from the precede studies. We considered that the microbes came mainly from the Zunil crater on Mars. We found that 70-80% of the microbes are likely to be dispersed all over the moon surface and are rapidly sterilized due to radiation except for those microbes within a thick ejecta deposit produced by meteoroids. The other 20-30% might be shielded from radiation by thick regolith layers that formed at collapsed layers in craters produced by Mars rock impacts. The total number of potentially surviving microbes at the thick ejecta deposits is estimated to be 3-4 orders of magnitude lower than at the Mars rock craters. The microbe concentration is irregular in the horizontal direction and is largely depth-dependent due to the radiation sterilization. The surviving fraction of transported microbes would be only 1 ppm on Phobos and 100 ppm on Deimos, suggesting that the transport processes and radiation severely affect microbe survival. The microbe sampling probability from the Martian moons was also investigated. We suggest that sample return missions from the Martian moons are classified into Unrestricted Earth-Return missions for 30 g samples and 10 cm depth sampling, even in our conservative scenario. We also conducted a full statistical analysis for sampling the regolith of Phobos to include the effects of uncertainties in input parameters on the sampling probability. The most likely probability of microbial contamination for return samples is estimated to be two orders of magnitude lower than the criterion defined by the planetary protection policy of the Committee on Space Research (COSPAR)., Comment: 50 pages, 16 figures, 2 tables, The abstract was shortened to meet the arXiv words limit, accepted for publication in Life Sciences in Space Research, special issues on planetary protection

Potential microbial contamination of Martian moons, Phobos and Deimos, which can be brought about by transportation of Mars ejecta produced by meteoroid impacts on the Martian surface, has been comprehensively assessed in a statistical approach, based on the most probable history of recent major gigantic meteoroid collisions on the Martian surface. This article is the first part of our study to assess potential microbial density in Mars ejecta departing from the Martian atmosphere, as a source of the second part where statistical analysis of microbial contamination probability is conducted. Potential microbial density on the Martian surface as the source of microorganisms was estimated by analogy to the terrestrial areas having the similar arid and cold environments, from which a probabilistic function was deduced as the asymptotic limit. Microbial survival rate during hypervelocity meteoroid collisions was estimated by numerical analysis of impact phenomena with and without taking internal friction and plastic deformation of the colliding meteoroid and the target ground into consideration. Trajectory calculations of departing ejecta through the Martian atmosphere were conducted with taking account of aerodynamic deceleration and heating by the aid of computational fluid dynamic analysis. It is found that Mars ejecta smaller than 0.03 m in diameter hardly reach the Phobos orbit due to aerodynamic deceleration, or mostly sterilized due to significant aerodynamic heating even though they can reach the Phobos orbit and beyond. Finally, the baseline dataset of microbial density in Mars ejecta departing for Martian moons has been presented for the second part of our study., Comment: 24 pages, 12 figures, and 5 tables, accepted for publication in Life Sciences in Space Research, special issues on planetary protection

This paper aims to make a graph representing an essential skeleton of a character from an image that includes a machine printed or a handwritten character using growing neural gas (GNG) method and relative network graph (RNG) algorithm. The visual system in our brain can recognize printed characters and handwritten characters easily, robustly, and precisely. How does our brain robustly recognize characters? The visual processing in our brain uses the essential features of an object, such as crosses and corners. These features will be helpful for character recognition by a computer. However, extraction of the features is difficult. If the skeleton of a character is represented as a graph, we can more easily extract the features. To extract the skeleton of a character as a graph from an image, this paper proposes the new approach using GNG and RNG algorithm. I achieved to extract skeleton graphs from images including distorted, noisy, and handwritten characters.

HAYABUSA is the first spacecraft ever to land on and lift off from any celestial body other than the moon. The mission, which returned asteroid samples to the Earth while overcoming various technical hurdles, ended on June 13, 2010, with the planned atmospheric re-entry. In order to safely deliver the sample return capsule, the HAYABUSA spacecraft ended its 7-year journey in a brilliant "artificial fireball" over the Australian desert. Spectroscopic observation was carried out in the near-ultraviolet and visible wavelengths between 3000 and 7500 \AA at 3 - 20 \AA resolution. Approximately 100 atomic lines such as Fe I, Mg I, Na I, Al I, Cr I, Mn I, Ni I, Ti I, Li I, Zn I, O I, and N I were identified from the spacecraft. Exotic atoms such as Cu I, Mo I, Xe I and Hg I were also detected. A strong Li I line (6708 \AA) at a height of ~55 km originated from the onboard Li-Ion batteries. The FeO molecule bands at a height of ~63 km were probably formed in the wake of the spacecraft. The effective excitation temperature as determined from the atomic lines varied from 4500 K to 6000 K. The observed number density of Fe I was about 10 times more abundant than Mg I after the spacecraft explosion. N2+(1-) bands from a shock layer and CN violet bands from the sample return capsule's ablating heat shield were dominant molecular bands in the near-ultraviolet region of 3000 - 4000 \AA. OH(A-X) band was likely to exist around 3092 \AA. A strong shock layer from the HAYABUSA spacecraft was rapidly formed at heights between 93 km and 83 km, which was confirmed by detection of N2+(1-) bands with a vibration temperature of ~13000 K. Gray-body temperature of the capsule at a height of ~42 km was estimated to be ~2437 K which is matched to a theoretical prediction. The final message of the HAYABUSA spacecraft and its sample return capsule are discussed through our spectroscopy., Comment: Accepted for publication in PASJ, 22 pages, 7 figures, 6 tables