Aurelia Butler

Appropriately balancing a videogame’s difficulty is essential to provide players with a pleasant experience. In a recent study, Korean scientists developed a novel approach for dynamic difficulty adjustment where the players’ emotions are estimated using in-game data, and the difficulty level is tweaked accordingly to maximize player satisfaction. Their efforts could contribute to balancing the difficulty of games and making them more appealing to all types of players.
Difficulty is a tough aspect to balance in video games. Some people prefer videogames that present a challenge whereas others enjoy an easy experience. To make this process easier, most developers use ‘dynamic difficulty adjustment (DDA).’ The idea of DDA is to adjust the difficulty of a game in real time according to player performance. For example, if player performance exceeds the developer’s expectations for a given difficulty level, the game’s DDA agent can automatically raise the difficulty to increase the challenge presented to the player. Though useful, this strategy is limited in that only player performance is taken into account, not how much fun they are actually having.
In a recent study published in Expert Systems With Applications, a research team from the Gwangju Institute of Science and Technology in Korea decided to put a twist on the DDA approach. Instead of focusing on the player’s performance, they developed DDA agents that adjusted the game’s difficulty to maximize one of four different aspects related to a player’s satisfaction: challenge, competence, flow, and valence. The DDA agents were trained via machine learning using data gathered from actual human players, who played a fighting game against various artificial intelligences (AIs) and then answered a questionnaire about their experience.
Using an algorithm called Monte-Carlo tree search, each DDA agent employed actual game data and simulated data to tune the opposing AI’s fighting style in a way that maximized a specific emotion, or ‘affective state.’ “One advantage of our approach over other emotion-centered methods is that it does not rely on external sensors, such as electroencephalography,” comments Associate Professor Kyung-Joong Kim, who led the study. “Once trained, our model can estimate player states using in-game features only.”
The team verified — through an experiment with 20 volunteers — that the proposed DDA agents could produce AIs that improved the players’ overall experience, no matter their preference. This marks the first time that affective states are incorporated directly into DDA agents, which could be useful for commercial games. “Commercial game companies already have huge amounts of player data. They can exploit these data to model the players and solve various issues related to game balancing using our approach,” remarks Associate Professor Kim. Worth noting is that this technique also has potential for other fields that can be ‘gamified,’ such as healthcare, exercise, and education.
This paper was made available online on June 3, 2022, and will be published in Volume 205 of the journal on November 1, 2022.
Story Source:
Materials provided by GIST (Gwangju Institute of Science and Technology). Note: Content may be edited for style and length. More

Over the past few decades, very few new antibiotics have been developed, largely because current methods for screening potential drugs are prohibitively expensive and time-consuming. One promising new strategy is to use computational models, which offer a potentially faster and cheaper way to identify new drugs.
A new study from MIT reveals the potential and limitations of one such computational approach. Using protein structures generated by an artificial intelligence program called AlphaFold, the researchers explored whether existing models could accurately predict the interactions between bacterial proteins and antibacterial compounds. If so, then researchers could begin to use this type of modeling to do large-scale screens for new compounds that target previously untargeted proteins. This would enable the development of antibiotics with unprecedented mechanisms of action, a task essential to addressing the antibiotic resistance crisis.
However, the researchers, led by James Collins, the Termeer Professor of Medical Engineering and Science in MIT’s Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering, found that these existing models did not perform well for this purpose. In fact, their predictions performed little better than chance.
“Breakthroughs such as AlphaFold are expanding the possibilities for in silico drug discovery efforts, but these developments need to be coupled with additional advances in other aspects of modeling that are part of drug discovery efforts,” Collins says. “Our study speaks to both the current abilities and the current limitations of computational platforms for drug discovery.”
In their new study, the researchers were able to improve the performance of these types of models, known as molecular docking simulations, by applying machine-learning techniques to refine the results. However, more improvement will be necessary to fully take advantage of the protein structures provided by AlphaFold, the researchers say.
Collins is the senior author of the study, which appears today in the journal Molecular Systems Biology. MIT postdocs Felix Wong and Aarti Krishnan are the lead authors of the paper. More

The sense of smell is one of the basic senses of animal species. It is critical to finding food, realizing attraction, and sensing danger. Humans detect smells, or odorants, with olfactory receptors expressed in olfactory nerve cells. These olfactory impressions of odorants on nerve cells are associated with their molecular features and physicochemical properties. This makes it possible to tailor odors to create an intended odor impression. Current methods only predict olfactory impressions from the physicochemical features of odorants. But, that method cannot predict the sensing data, which is indispensable for creating smells.
To tackle this issue, scientists from Tokyo Institute of Technology (Tokyo Tech) have employed the innovative strategy of solving the inverse problem. Instead of predicting the smell from molecular data, this method predicts molecular features based on the odor impression. This is achieved using standard mass spectrum data and machine learning (ML) models. “We used a machine-learning-based odor predictive model that we had previously developed to obtain the odor impression. Then we predicted the mass spectrum from odor impression inversely based on the previously developed forward model,” explains Professor Takamichi Nakamoto, the leader of the research effort by Tokyo Tech. The findings have been published in PLoS One.
The mass spectra of odor mixtures is obtained by a linear combination of the mass spectra of single components. This simple method allows for the quick preparation of the predicted spectra of odor mixtures and can also predict the required mixing ratio, an important part of the recipe for new odor preparation. “For example, we show which molecules give the mass spectrum of apple flavor with enhanced ‘fruit’ and ‘sweet’ impressions. Our analysis method shows that combinations of either 59 or 60 molecules give the same mass spectrum as the one obtained from the specified odor impression. With this information, and the correct mixing ratio needed for a certain impression, we could theoretically prepare the desired scent,” highlights Prof. Nakamoto.
This novel method described in this study can provide highly accurate predictions of the physicochemical properties of odor mixtures, as well as the mixing ratios required to prepare them, thereby opening the door to endless tailor-made fragrances.
Story Source:
Materials provided by Tokyo Institute of Technology. Note: Content may be edited for style and length. More

An international team led by researchers at the RIKEN Cluster for Pioneering Research (CPR) has engineered a system for creating remote controlled cyborg cockroaches, equipped with a tiny wireless control module that is powered by a rechargeable battery attached to a solar cell. Despite the mechanic devices, ultrathin electronics and flexible materials allow the insects to move freely. These achievements, reported in the scientific journal npj Flexible Electronics on September 5, will help make the use of cyborg insects a practical reality.
Researchers have been trying to design cyborg insects — part insect, part machine — to help inspect hazardous areas or monitor the environment. However, for the use of cyborg insects to be practical, handlers must be able to control them remotely for long periods of time. This requires wireless control of their leg segments, powered by a tiny rechargeable battery. Keeping the battery adequately charged is fundamental — nobody wants a suddenly out-of-control team of cyborg cockroaches roaming around. While it’s possible to build docking stations for recharging the battery, the need to return and recharge could disrupt time-sensitive missions. Therefore, the best solution is to include an on-board solar cell that can continuously ensure that the battery stays charged.
All of this is easier said than done. To successfully integrate these devices into a cockroach that has limited surface area required the research team to develop a special backpack, ultrathin organic solar cell modules, and an adhesion system that keeps the machinery attached for long periods of time while also allowing natural movements.
Led by Kenjiro Fukuda, RIKEN CPR, the team experimented with Madagascar cockroaches, which are approximately 6 cm long. They attached the wireless leg-control module and lithium polymer battery to the top of the insect on the thorax using a specially designed backpack, which was modeled after the body of a model cockroach. The backpack was 3D printed with an elastic polymer and conformed perfectly to the curved surface of the cockroach, allowing the rigid electronic device to be stably mounted on the thorax for more than a month.
The ultrathin 0.004 mm thick organic solar cell module was mounted on the dorsal side of the abdomen. “The body-mounted ultrathin organic solar cell module achieves a power output of 17.2 mW, which is more than 50 times larger than the power output of current state-of-the art energy harvesting devices on living insects,” according to Fukuda.
The ultrathin and flexible organic solar cell, and how it was attached to the insect, proved necessary to ensure freedom of movement. After carefully examining natural cockroach movements, the researchers realized that the abdomen changes shape and portions of the exoskeleton overlap. To accommodate this, they interleaved adhesive and non-adhesive sections onto the films, which allowed them to bend but also stay attached. When thicker solar cell films were tested, or when the films were uniformly attached, the cockroaches took twice as long to run the same distance, and had difficulty righting themselves when on their backs.
Once these components were integrated into the cockroaches, along with wires that stimulate the leg segments, the new cyborgs were tested. The battery was charged with pseudo-sunlight for 30 minutes, and animals were made to turn left and right using the wireless remote control.
“Considering the deformation of the thorax and abdomen during basic locomotion, a hybrid electronic system of rigid and flexible elements in the thorax and ultrasoft devices in the abdomen appears to be an effective design for cyborg cockroaches,” says Fukuda. “Moreover, since abdominal deformation is not unique to cockroaches, our strategy can be adapted to other insects like beetles, or perhaps even flying insects like cicadas in the future.”
Story Source:
Materials provided by RIKEN. Note: Content may be edited for style and length. More

Monitoring and measuring forest ecosystems is a complex challenge because of an existing combination of softwares, collection systems and computing environments that require increasing amounts of energy to power. The University of Maine’s Wireless Sensor Networks (WiSe-Net) laboratory has developed a novel method of using artificial intelligence and machine learning to make monitoring soil moisture more energy and cost efficient — one that could be used to make measuring more efficient across the broad forest ecosystems of Maine and beyond.
Soil moisture is an important variable in forested and agricultural ecosystems alike, particularly under the recent drought conditions of past Maine summers. Despite the robust soil moisture monitoring networks and large, freely available databases, the cost of commercial soil moisture sensors and the power that they use to run can be prohibitive for researchers, foresters, farmers and others tracking the health of the land.
Along with researchers at the University of New Hampshire and University of Vermont, UMaine’s WiSe-Net designed a wireless sensor network that uses artificial intelligence to learn how to be more power efficient in monitoring soil moisture and processing the data. The research was funded by a grant from the National Science Foundation.
“AI can learn from the environment, predict the wireless link quality and incoming solar energy to efficiently use limited energy and make a robust low cost network run longer and more reliably,” says Ali Abedi, principal investigator of the recent study and professor of electrical and computer engineering at the University of Maine.
The software learns over time how to make the best use of available network resources, which helps produce power efficient systems at a lower cost for large scale monitoring compared to the existing industry standards.
WiSe-Net also collaborated with Aaron Weiskittel, director of the Center for Research on Sustainable Forests, to ensure that all hardware and software research is informed by the science and tailored to the research needs.
“Soil moisture is a primary driver of tree growth, but it changes rapidly, both daily as well as seasonally,” Weiskittel says. “We have lacked the ability to monitor effectively at scale. Historically, we used expensive sensors that collected at fixed intervals — every minute, for example — but were not very reliable. A cheaper and more robust sensor with wireless capabilities like this really opens the door for future applications for researchers and practitioners alike.”
The study was published Aug. 9, 2022, in the Springer’s International Journal of Wireless Information Networks.
Although the system designed by the researchers focuses on soil moisture, the same methodology could be extended to other types of sensors, like ambient temperature, snow depth and more, as well as scaling up the networks with more sensor nodes.
“Real-time monitoring of different variables requires different sampling rates and power levels. An AI agent can learn these and adjust the data collection and transmission frequency accordingly rather than sampling and sending every single data point, which is not as efficient,” Abedi says.
Story Source:
Materials provided by University of Maine. Note: Content may be edited for style and length. More

An ingenious approach toward developing low-power, high-speed, and high-density memory devices is based on spintronics, an emerging frontier in technology that harnesses a degree of freedom of electrons known as “spin.” Put simply, electrons, along with their negative charge, possess a “spin” whose orientation can be controlled using magnetic fields. This is particularly relevant for magnetic insulators, in which the electrons cannot move around, but the “spin” remains controllable. In these materials, the magnetic excitations can give rise to a “spin current,” which forms the basis of spintronics.
Scientists have been looking for efficient methods to generate the spin current. The “photogalvanic effect,” a phenomenon characterized by the generation of dc current from light illumination, is particularly useful in this regard. Studies have found that a “photogalvanic” spin current can be generated similarly using the magnetic fields in electromagnetic waves. However, we currently lack candidate materials and a general mathematical formulation for exploring this phenomenon.
Now, Associate Professor Hiroaki Ishizuka from Tokyo Institute of Technology (Tokyo Tech), along with his colleague, has addressed these issues. In their recent breakthrough published in Physical Review Letters, they presented a general formula that can be used to calculate the photogalvanic spin current induced by transverse oscillating magnetic excitations. They then used this formula to understand how photogalvanic spin currents arise in bilayer chromium (Cr) trihalide compounds, namely chromium triiodide (CrI3) and chromium tribromide (CrBr3).
“Unlike past studies that considered longitudinal oscillating magnetic fields for generating spin currents, our study focuses on transverse oscillating magnetic fields. Based on this, we found that processes involving one magnon (quantum of spin wave excitations) band as well as two magnon bands contribute to the spin current,” elaborates Dr. Ishizuka, speaking of their findings.
Using their formula, the duo found that both CrI3 and CrBr3 showed a large photogalvanic spin current for magnetic excitations corresponding to electromagnetic waves at gigahertz and terahertz frequencies. However, the current only appeared when the spins showed antiferromagnetic ordering, i.e., successive spins were anti-parallel, as opposed to ferromagnetic ordering (successive spins were parallel). Moreover, the spin current direction was governed by the orientation of the antiferromagnetic ordering (whether the spins on the first and second layers were arranged up-down or down-up). Additionally, they pointed out that, unlike previous findings that attributed the spin current to only the two-magnon process, their formula showed that a large response was, in general, possible with the single magnon process.
These results suggest that bilayer CrI3 and CrBr3 are strong candidates for investigating the mechanism associated with photogalvanic spin current generation. “Our study not only predicts unforeseen contributions to the spin current but also provides a guideline for the design of novel materials driven by the photogalvanic effect of magnetic excitations,” highlights Dr. Ishizuka.
Story Source:
Materials provided by Tokyo Institute of Technology. Note: Content may be edited for style and length. More

Be it magnets or superconductors: materials are known for their various properties. However, these properties may change spontaneously under extreme conditions. Researchers at the Technische Universität Dresden (TUD) and the Technische Universität München (TUM) have discovered an entirely new type of such phase transitions. They display the phenomenon of quantum entanglement involving many atoms, which previously has only been observed in the realm of few atoms. The results were recently published in the scientific journal Nature.
In physics, Schroedinger’s cat is an allegory for two of the most awe-inspiring effects of quantum mechanics: entanglement and superposition. Researchers from Dresden and Munich have now observed these behaviors on a much larger scale than that of the smallest of particles. Until now, materials that display properties like, e.g., magnetism have been known to have so-called domains — islands in which the materials properties are homogeneously either of one or a different kind (imagine them being either black or white, for example). Looking at lithium holmium fluoride (LiHoF4), the physicists have now discovered a completely new phase transition, at which the domains surprisingly exhibit quantum mechanical features, resulting in their properties becoming entangled (being black and white at the same time). “Our quantum cat now has a new fur because we’ve discovered a new quantum phase transition in LiHoF4 which has not previously been known to exist,” comments Matthias Vojta, Chair of Theoretical Solid State Physics at TUD.
Phase transitions and entanglement
We can easily observe the spontaneously changing properties of a substance if we look at water: at 100 degrees Celsius it evaporates into a gas, at zero degrees Celsius it freezes into ice. In both cases, these new states of matter form as a consequence of a phase transition where the water molecules rearrange themselves, thus changing the characteristics of the matter. Properties like magnetism or superconductivity emerge as a result of electrons undergoing phase transitions in crystals. For phase transitions at temperatures approaching the absolute zero at -273.15 degrees Celsius, quantum mechanical effects such as entanglement come into play, and one speaks of quantum phase transitions. “Even though there are more than 30 years of extensive research dedicated to phase transitions in quantum materials, we had previously assumed that the phenomenon of entanglement played a role only on a microscopic scale, where it involves only a few atoms at a time,” explains Christian Pfleiderer, Professor of Topology of Correlated Systems at the TUM.
Quantum entanglement is one of the most astonishing phenomena of physics, where the entangled quantum particles exist in a shared superposition state that allows for usually mutually exclusive properties (e.g., black and white) to occur simultaneously. As a rule, the laws of quantum mechanics only apply to microscopic particles. The research teams from Munich and Dresden have now succeeded in observing effects of quantum entanglement on a much larger scale, that of thousands of atoms. For this, they have chosen to work with the well-known compound LiHoF4.
Spherical samples enable precision measurements
At very low temperatures, LiHoF4 acts as a ferromagnet where all magnetic moments spontaneously point in the same direction. If you then apply a magnetic field exactly vertically to the preferred magnetic direction, the magnetic moments will change direction, which is known as fluctuations. The higher the magnetic field strength, the stronger these fluctuations become, until, eventually, the ferromagnetism disappears completely at a quantum phase transition. This leads to the entanglement of neighboring magnetic moments. “If you hold up a LiHoF4 sample to a very strong magnet, it suddenly ceases to be spontaneously magnetic. This has been known for 25 years,” summarizes Vojta.
What is new is what happens when you change the direction of the magnetic field. “We discovered that the quantum phase transition continues to occur, whereas it had previously been believed that even the smallest tilt of the magnetic field would immediately suppress it,” explains Pfleiderer. Under these conditions, however, it is not individual magnetic moments but rather extensive magnetic areas, so-called ferromagnetic domains, that undergo these quantum phase transitions. The domains constitute entire islands of magnetic moments pointing in the same direction. “We have used spherical samples for our precision measurements. That is what enabled us to precisely study the behavior upon small changes in the direction of the magnetic field,” adds Andreas Wendl, who conducted the experiments as part of his doctoral dissertation.
From fundamental physics to applications
“We have discovered an entirely new type of quantum phase transitions where entanglement takes place on the scale of many thousands of atoms instead of just in the microcosm of only a few,” explains Vojta. “If you imagine the magnetic domains as a black-and-white pattern, the new phase transition leads to either the white or the black areas becoming infinitesimally small, i.e., creating a quantum pattern, bevor dissolving completely.” A newly developed theoretical model successfully explains the data obtained from the experiments. “For our analysis, we generalized existing microscopic models and also took into account the feedback of the large ferromagnetic domains to the microscopic properties,” elaborates Heike Eisenlohr, who performed the calculations as part of her PhD thesis.
The discovery of the new quantum phase transitions is important as a foundation and general frame of reference for the research of quantum phenomena in materials, as well as for new applications. “Quantum entanglement is applied and used in technologies like quantum sensors and quantum computers, amongst other things,” says Vojta. Pfleiderer adds: “Our work is in the area of fundamental research, which, however, can have a direct impact on the development of practical applications, if you use the materials properties in a controlled way.”
The research has been financially supported by the Excellence Strategy of the German Federal and State Governments within the Würzburg-Dresden Cluster of Excellence Complexity and Topology in Quantum Matter (ct.qmat) and the Cluster of Excellence Munich Center for Quantum Science and Technology (MCQST). In addition, the work has been supported by the European Research Council (ERC) via the Advanced Grant ExQuiSid and by the Deutsche Forschungsgemeinschaft (DFG) within the Collaborative Research Centers (SFB) 1143 und TRR80. More

Artificial intelligence (AI) can be trained to recognise whether a tissue image contains a tumour. However, exactly how it makes its decision has remained a mystery until now. A team from the Research Center for Protein Diagnostics (PRODI) at Ruhr-Universität Bochum is developing a new approach that will render an AI’s decision transparent and thus trustworthy. The researchers led by Professor Axel Mosig describe the approach in the journal Medical Image Analysis, published online on 24 August 2022.
For the study, bioinformatics scientist Axel Mosig cooperated with Professor Andrea Tannapfel, head of the Institute of Pathology, oncologist Professor Anke Reinacher-Schick from the Ruhr-Universität’s St. Josef Hospital, and biophysicist and PRODI founding director Professor Klaus Gerwert. The group developed a neural network, i.e. an AI, that can classify whether a tissue sample contains tumour or not. To this end, they fed the AI a large number of microscopic tissue images, some of which contained tumours, while others were tumour-free.
“Neural networks are initially a black box: it’s unclear which identifying features a network learns from the training data,” explains Axel Mosig. Unlike human experts, they lack the ability to explain their decisions. “However, for medical applications in particular, it’s important that the AI is capable of explanation and thus trustworthy,” adds bioinformatics scientist David Schuhmacher, who collaborated on the study.
AI is based on falsifiable hypotheses
The Bochum team’s explainable AI is therefore based on the only kind of meaningful statements known to science: on falsifiable hypotheses. If a hypothesis is false, this fact must be demonstrable through an experiment. Artificial intelligence usually follows the principle of inductive reasoning: using concrete observations, i.e. the training data, the AI creates a general model on the basis of which it evaluates all further observations.
The underlying problem had been described by philosopher David Hume 250 years ago and can be easily illustrated: No matter how many white swans we observe, we could never conclude from this data that all swans are white and that no black swans exist whatsoever. Science therefore makes use of so-called deductive logic. In this approach, a general hypothesis is the starting point. For example, the hypothesis that all swans are white is falsified when a black swan is spotted.
Activation map shows where the tumour is detected
“At first glance, inductive AI and the deductive scientific method seem almost incompatible,” says Stephanie Schörner, a physicist who likewise contributed to the study. But the researchers found a way. Their novel neural network not only provides a classification of whether a tissue sample contains a tumour or is tumour-free, it also generates an activation map of the microscopic tissue image.
The activation map is based on a falsifiable hypothesis, namely that the activation derived from the neural network corresponds exactly to the tumour regions in the sample. Site-specific molecular methods can be used to test this hypothesis.
“Thanks to the interdisciplinary structures at PRODI, we have the best prerequisites for incorporating the hypothesis-based approach into the development of trustworthy biomarker AI in the future, for example to be able to distinguish between certain therapy-relevant tumour subtypes,” concludes Axel Mosig.
Story Source:
Materials provided by Ruhr-University Bochum. Note: Content may be edited for style and length. More