Search

Learning efficient and interpretable policies has been a challenging task in reinforcement learning (RL), particularly in the visual RL setting with complex scenes. While neural networks have achieved competitive performance, the resulting policies are often over-parameterized black boxes that are difficult to interpret and deploy efficiently. More recent SRL frameworks have shown that high-level domain-specific programming logic can be designed to handle both policy learning and symbolic planning. However, these approaches rely on coded primitives with little feature learning, and when applied to high-dimensional visual scenes, they can suffer from scalability issues and perform poorly when images have complex object interactions. To address these challenges, we propose Differentiable Symbolic Expression Search (DiffSES), a novel symbolic learning approach that discovers discrete symbolic policies using partially differentiable optimization. By using object-level abstractions instead of raw pixel-level inputs, DiffSES is able to leverage the simplicity and scalability advantages of symbolic expressions, while also incorporating the strengths of neural networks for feature learning and optimization. Our experiments demonstrate that DiffSES is able to generate symbolic policies that are simpler and more and scalable than state-of-the-art SRL methods, with a reduced amount of symbolic prior knowledge.

Robotic-assisted total knee replacement (RA-TKR) is a significant advancement in orthopedic surgery, but intra-operative decision-making remains challenging. Pre-operative imaging techniques, particularly CT scans, have gained momentum, providing insights into the patient's anatomy, improving implant positioning and alignment. However, further research is needed to explore their influence on RA-TKR planning and execution.

Osteoarthritis of the knee is a common problem in the elderly, leading to severe morbidity. Total Knee Arthroplasty (TKA) is a widely validated surgery to provide a remarkable extent of knee function and simultaneously alleviates pain for knee osteoarthritis (OA). It is clearly understood that precision of the rotational alignment and accuracy of the technique in the placement of the femoral component is a prerequisite for excellent and successful outcomes of TKA. Advanced technology has now allowed surgeons to understand patient-specific variabilities in anatomical reference landmarks and the relationship of component positioning in relation to the reference landmarks to achieve accurate gap balancing with minimal soft tissue release.14Robotic Arm Assisted-Total Knee Arthroplasty (RAA-TKA) is a semi-automated system that enables us in replicating the same. Using this technology, the bony resections, component positions, probable component sizing and gap balancing can be tentatively planned preoperatively with CT Scan Analysis and executed intraoperatively. Hence this study was undertaken to estimate the relationship between femoral component placement to normal rotational landmarks such as the Posterior Condylar Axis (PCA) and to quantitatively evaluate coronal and sagittal plane correction obtained. Also, we aimed to use the data to detect any anatomical variations in the study population and evaluate the accuracy of predicted component sizing, including gender-based evaluation.

Purpose: In curriculum learning, the idea is to train on easier samples first and gradually increase the difficulty, while in self-paced learning, a pacing function defines the speed to adapt the training progress. While both methods heavily rely on the ability to score the difficulty of data samples, an optimal scoring function is still under exploration. Methodology: Distillation is a knowledge transfer approach where a teacher network guides a student network by feeding a sequence of random samples. We argue that guiding student networks with an efficient curriculum strategy can improve model generalization and robustness. For this purpose, we design an uncertainty-based paced curriculum learning in self-distillation for medical image segmentation. We fuse the prediction uncertainty and annotation boundary uncertainty to develop a novel paced-curriculum distillation (P-CD). We utilize the teacher model to obtain prediction uncertainty and spatially varying label smoothing with Gaussian kernel to generate segmentation boundary uncertainty from the annotation. We also investigate the robustness of our method by applying various types and severity of image perturbation and corruption. Results: The proposed technique is validated on two medical datasets of breast ultrasound image segmentation and robot-assisted surgical scene segmentation and achieved significantly better performance in terms of segmentation and robustness. Conclusion: P-CD improves the performance and obtains better generalization and robustness over the dataset shift. While curriculum learning requires extensive tuning of hyper-parameters for pacing function, the level of performance improvement suppresses this limitation.

Pancreatic cancer is a leading cause of death from cancer but only a minority of patients with pancreatic ductal adenocarcinomas are eligible for curative resection. The increasing role of neoadjuvant therapy provides hope of improving outcomes. However, progress is also reliant on advances in imaging that can identify disease earlier and accurately assess treatment response. Computed tomography remains the cornerstone in evaluation of resectability, offering excellent spatial resolution. However, in high-risk patients, additional magnetic resonance imaging and positron emission tomography-computed tomography may further guide treatment decisions. Conventional computed tomography can be limited in its ability to determine disease response after neoadjuvant therapy. Dual-energy computed tomography and computed tomography or magnetic resonance imaging perfusion studies emerging as potentially better alternatives. Combined with pioneering advances in radiomic analysis, these modalities also show promise in analysing tumour heterogeneity and thereby more accurately predicting outcomes. This article reviews these imaging techniques.

In this study, we report the complete genome sequencing of the Duck plague virus from India for the first time. The sequencing was done on the MinION nanopore sequencer from Oxford Nanopore Technologies. The closest relative is the European strain 2085v, with 99.98 and 99.8% identity at the amino acid and nucleotide level respectively. Moreover, 72 out of 77 ORFs are completely conserved between the 2 strains. The high similarity with the European strain over the only three other pathogenic strains reported from China points to the circulation of European strain in India. The fly pathways of migratory birds and co-habitation with native species being a probable reason. More complete genome data from diverse sampling locations are needed to characterize the genomic features, develop diagnostics, vaccines, and understand the evolution of the virus.

We report herewith tryptophan (Trp)-conjugated peptidomimetics that show intramolecular through-space association between the Trp units. Our investigation revealed that the proximal placement of Trp can lead to the emergence of a new and unanticipated fluorescent entity constituting a Trp–Trp dimer. Proton-induced modulation of fluorescence is a consequence of this work. Investigations with control compounds unequivocally revealed that the fluorescence property is not originated from the localized excited state but from the unprecedented Trp–Trp intramolecular dimer in the ground state itself. The present findings will initiate the biophysical scientists to have a relook at the fluorescence properties of Trp-containing proteins.

Hypoxia and inflammatory cytokines like interleukin-6 (IL-6, IL6) are strongly linked to cancer progression, and signal in part through the transcription factor Ccaat/enhancer-binding protein d (C/EBPd, CEBPD), which has been shown to promote mesenchymal features and malignant progression of glioblastoma. Here we report a different role for C/EBPd in breast cancer. We found that the C/EBPd protein is expressed in normal breast epithelial cells and in low-grade cancers. C/EBPd protein (but not mRNA) expression correlates with estrogen receptor (ER+) and progesterone receptor (PGR) expression and longer progression-free survival of breast cancer patients. Specifically in ER+ breast cancers, CEBPD—but not the related CEBPB—mRNA in combination with IL6correlated with lower risk of progression. Functional studies in cell lines showed that ERa promotes C/EBPd expression at the level of protein stability by inhibition of the FBXW7 pathway. Furthermore, we found that C/EBPd attenuates cell growth, motility and invasiveness by inhibiting expression of the SNAI2 (Slug) transcriptional repressor, which leads to expression of the cyclin-dependent kinase inhibitor CDKN1A (p21CIP1/WAF1). These findings identify a molecular mechanism by which ERa signaling reduces the aggressiveness of cancer cells, and demonstrate that C/EBPd can have different functions in different types of cancer. Furthermore, our results support a potentially beneficial role for the IL-6 pathway specifically in ER+ breast cancer and call for further evaluation of the role of intra-tumoral IL-6 expression and of which cancers might benefit from current attempts to target the IL-6 pathway as a therapeutic strategy.

AbstractThe purpose of this paper is to introduce and study the concepts of homomorphism, fuzzy multiset transformation semigroup and coverings of fuzzy multiset finite automata. We establish their basic properties for the algebraic study of fuzzy multiset automata theory.

Unique information about the dynamo process acting at Jupiter can be inferred by modeling and interpreting its magnetic field. Using the fluxgate magnetometer measurements acquired during the 4 years of the Juno mission, we derive a magnetic field model which describes simultaneously the main field and the secular variation (SV) up to spherical harmonic degrees 16 and 8, respectively. Apart from the Earth's, this is the first time another planetary magnetic field along with its time variation is described to such a high degree. We use properties of the power spectrum of the static field to infer the upper boundary of the dynamo convective region at 0.830 ± 0.022 Jupiter radius. The SV and correlation times are relatively comparable to the Earth's and indicate that the field is dominated by advection. The field and SV morphologies suggest zonal as well as non‐zonal deep fluid motions. The interior of Jupiter can be described broadly as a dense core surrounded by fluids, dominantly hydrogen and helium. The hydrogen‐rich metallic fluid generates the strongest planetary magnetic field in the Solar System. Modeling and interpreting this field gives essential information about the dynamo process inside Jupiter. We use the Juno mission data throughout 4 years (or, 28 orbits) to derive an internal magnetic field and secular variation (SV) model using spherical harmonic functions. We compute a magnetic field model to degree 16 for its static part, and model its temporal variation to degree 8. The power spectrum of the magnetic field model is used to investigate the radius of the dynamo region. We infer that the convective region has an upper boundary at 0.830 ± 0.022 Jupiter radius. The strength of the annual change of field is relatively comparable to the Earth's. The slope of the SV timescales indicates that the dynamo is dominated by advective effects. The SV displays a maximum near the equator with a bi‐polar structure in agreement with zonal drift of the Great Blue Spot. However, numerous small scale SV structures suggest that the flow at the interior is complex involving both zonal and non‐zonal features. Magnetic field of Jupiter is modeled from Juno's first 4 years of observationsA degree 16 magnetic field model and degree 8 secular variation model are derivedThe model indicates a dynamo not far from the surface and complex motions deep inside Jupiter Magnetic field of Jupiter is modeled from Juno's first 4 years of observations A degree 16 magnetic field model and degree 8 secular variation model are derived The model indicates a dynamo not far from the surface and complex motions deep inside Jupiter

Abstract: Defects such as dislocations and interfaces play a crucial role in the performance of heterostructure devices. The full potential of GaAs on Si heterostructures can only be realized by controlling the defect density. The reduction of threading dislocations by the use of strained layer superlattices has been studied in these heterostructures. Several superlattice structures have been used to reduce the density of threading dislocations in the GaAs epilayer. In this study, we have optimized the use of strained layer superlattices with respect to the position, period and number to reduce and control the dislocation density. The use of strained layer superlattices in conjunction with rapid thermal annealing was found to be a most effective method for reducing the threading dislocation density. Transmission electron microscopy has been used to study the dislocation density reduction and the interaction of threading dislocations with the strained layers. A model has been developed based on energy considerations to determine the critical thickness required for the bending of threading dislocations.

Abstract: We have investigated the formation of TiSi2 and CoSi2 thin films on Si(100) substrates using laser (wave length 248 nm, pulse duration 40 ns and repetition rate 5 Hz) physical vapor deposition (LPVD). The films were deposited from solid targets of TiSi2 and CoSi2 in vacuum with the substrate temperature optimized at 600� C. The films were characterized using x-ray diffraction, scanning electron microscopy (SEM), transmission electron microscopy (TEM) and four point probe ac resistivity. The films were found to be polycrystalline with a texture. The room temperature resistivity was found to be 16 μΩ-@#@ cm and 23 μΩ-cm for TiSi2 and CoSi2 films, respectively. We optimized the processing parameters so as to get particulate free surface. TEM results show that the silicide/silicon interface is quite smooth and there is no perceptible interdiffusion across the interface.

Defects such as dislocations and interfaces play a crucial role in the performance of heterostructure devices. The full potential of GaAs on Si heterostructures can only be realized by controlling the defect density. The nature of misfit dislocations at the heterointerface has been studied and a mechanism for the formation of 60° and 90° misfit dislocations has been proposed. Threading dislocations in the epilayer are the most prominent defects and their density has to be controlled to fully utilize the properties of semiconductor heterostructures. Various processes to reduce defect densities in the epilayers have been discussed and in particular, the use of strained layer superlattices to reduce the threading dislocation density has been presented in this paper. Several superlattice structures have been used to reduce the density of threading dislocations in the GaAs epilayer. In this study, we have optimized the use of strained layer superlattices with respect to the position, period and number to reduce and control the dislocation density. The use of strained layer superlattices in conjunction with rapid thermal annealing was found to be the most effective method in reducing threading dislocation density. Transmission electron microscopy has been used to study the dislocation density reduction and the interaction of threading dislocations with the strained layers.

Zusammenfassung Der Einfluß eines Magnetfeldes auf den Widerstand von Nickel wurde mit einer Wechselstrombrücke untersucht. Es zeigte sich, daß die Erscheinung hier sich insofern vom Gleichstromwiderstand unterscheidet, als die Hysteresis geringer ist, als überhaupt kein remanenter Widerstand auftritt und als die Widerstandskurve in Abhängigkeit vom Magnetfeld eine andere Form hat. Die Ergebnisse erweisen eine enge Beziehung zwischen der Widerstandsänderung und der Magnetisierung der Substanz. Die mit Wechselstrom aufgenommene Widerstandskurve in Abhängigkeit vom Magnetfeld ist von vielen Unregelmäßigkeiten frei, die oft mit Gleichstrom gefunden werden, und man kann annehmen, daß die mit dieser Methode gefundenen Daten für das Verständnis dieser Widerstandsänderung von großem Wert sein werden.

A novel mathematical model is presented for determination of creep behavior of Aluminum 6061 alloy subjected to constant-stress uniaxial tensile creep experiment. The methodology employed to obtain the constants is by conducting experiments at the temperature range of 573 K to 673 K (300degC to 400degC) at different stress intervals 64.5 MPa and 98.1 MPa. The creep properties such as creep strain rate for all three stages, rupture time, rupture stress, rupture strain and percentage elongation of specimen are calculated. Depending on the temperature and the applied stress, exhibits significant effects on the strain rate and activation energy. The results show that the creep behaviors of the studied aluminum alloy strongly depend on the creep temperature, external stress, and creep time.

IN the course of experiments on the latent splitting of bars as reported by one of us1, we noticed a very curious and interesting phenomenon in zinc.