Tashkent University of Information Technologies

Cryptocurrency trading presents significant challenges due to extreme market volatility, rapid regime transitions, and non-stationary dynamics that render traditional trading strategies ineffective. Existing reinforcement learning approaches for cryptocurrency trading typically employ simplistic profit-based reward functions that fail to adequately capture risk management considerations, market microstructure costs, temporal dependencies, and regime-specific optimal behaviors. This limitation often results in strategies that perform well during favorable market conditions but suffer catastrophic losses during downturns. This paper introduces five novel reward functions grounded in economic utility theory, market microstructure, behavioral finance, adaptive risk management, and regime-conditional optimization. We systematically evaluate these reward functions across three reinforcement learning algorithms (Deep Q-Network, Proximal Policy Optimization, and Advantage Actor–Critic) and four distinct market regimes (bull, bear, high volatility, and recovery), using Bitcoin hourly data from 2018–2022. Our comprehensive experimental evaluation demonstrates that the Adaptive Risk Control reward function achieves exceptional performance, with a Sharpe ratio of 2.47, cumulative return of 26.4%, and maximum drawdown of only 16.8% during the predominantly bearish 2022 test period. Critically, regime-specific analysis reveals substantial performance heterogeneity: Adaptive Risk Control excels during high volatility (Sharpe ratio 3.21), while Temporal Coherence and Asymmetric Market-Conditional rewards dominate in trending and bear markets, respectively. These findings establish that sophisticated, theory-grounded reward engineering—rather than algorithmic innovations alone—constitutes the primary lever for improving RL trading systems, enabling positive risk-adjusted returns even during severe market downturns.

This article presents a systematic review of modern 2D and 3D deep learning approaches used in the fields of Human Pose Estimation (HPE) and Skeleton-Based Action Recognition (SBAR). The research was conducted based on the SALSA methodology and the published works in leading scientific databases were analyzed. Within the framework of the review, approaches based on the detection of human joint points from image and video sequences, reconstruction of the skeletal structure, and modeling of actions in the spatial and spatiotemporal (ST) domains were compared. The article reviews the architectures of 2D and 3D HPE models, joint hiding in multi-person scenes, real-time requirements, and the possibilities of application in embedded devices. The spatial, spatiotemporal, and graphical features used in SBAR systems and their impact on computational complexity and energy efficiency are also analyzed. The performance of the models was compared based on evaluation criteria such as MPJPE (Mean Per Joint Position Error), AP (Average Precision), RMSE (Root Mean Square Error), and Pearson correlation. The results show that skeleton-based approaches are effective solutions for real-time and resource-constrained systems. However, choosing the optimal model and features requires a trade-off between accuracy, computational complexity, and energy efficiency. This paper identifies promising architectures, features, and hardware adaptation strategies for practical applications of HPE and SBAR systems in real-world environments.

We study transparent boundary conditions (TBCs) for the time-dependent heat equation in a branching, quasi-one-dimensional domain modeled as a metric star graph. By combining the classical concept of TBCs with the theory of partial differential equations (PDEs) on networks, we derive an exact vertex condition that ensures unobstructed thermal flow across junctions. Specifically,we derive a sum rule for the diffusion coefficients that eliminates thermal backflow at the vertex.We demonstrate the validity of this analytical model numerically using a Crank-Nicolson finite-difference method, which confirms smooth, unobstructed heat propagation through the network.These results provide a practical mathematical framework for tunable control and optimization of thermal diffusion in low-dimensional structures. In particular, combining the well-known concept of the TBCs and theory for PDE on metric graphs, we propose a mathematical model providing a control tool for thermal diffusion in networks.

Aerial scenes exhibit a mismatch between where computation is spent and where information resides: targets occupy a small, unevenly distributed fraction of the frame, yet detectors process every region at identical resolution. Tiling and clustering pipelines exploit this sparsity, but the rule deciding which regions deserve magnification is hand-designed and never observes the detection loss. AFWD-Net removes that separation. A coarse density estimator, run on a heavily downsampled copy of the frame at negligible cost, produces a spatial prior over likely target locations; Gumbel top-K relaxation and a spatial-transformer sampler convert this prior into a small set of high-resolution crops whose centers and scales are differentiable, so the gradient that measures detection quality also determines where resolution is allocated. Within each crop, a learnable wavelet decomposition separates approximation from directional detail sub-bands and enhances the latter under edge supervision confined to annotated regions, restoring the high-frequency structure that stride downsampling erases, precisely the cue separating a genuine sub-32-pixel target from background clutter. Detection queries are seeded from the same density field rather than a fixed lattice, and crop-level predictions are merged by density-calibrated soft suppression. On VisDrone2019-DET the model attains 49.3% mAP@0.5, 30.4% mAP@[0.5:0.95] and 22.8% APS at 15.7 M parameters; a distilled 7.2 M student remains competitive with all compared methods. Experiments on UAVDT, AI-TOD-v2 and SODA-A assess cross-dataset generalization.

Ultrafast laser nanostructuring of semiconductor substrates can markedly improve the sensitivity of Surface-Enhanced Raman Spectroscopy (SERS). In this work, we investigated the SERS response of rhodamine 6G (R6G) on regular laser-induced periodic surface structures (LIPSS) fabricated on silicon using femtosecond laser pulses and subsequently decorated with silver nanoparticles produced by femtosecond laser ablation in ethanol. The resulting hybrid substrates enabled uniform analyte adsorption and efficient formation of plasmonic hotspots, allowing reliable detection of R6G down to 10(-)(7) M. Compared to planar silicon, the nanostructured surfaces exhibited an approximately 25-fold higher enhancement factor, arising from stronger light backscattering by the periodic architecture and the tighter packing of Ag nanoparticles within the LIPSS valleys, which reduces interparticle spacing and increases electromagnetic coupling. FDTD simulations further revealed that the electric field intensifies dramatically when nanoparticle separations approach 2 nm significantly stronger than at 10 nm corroborating the experimentally observed enhancement.