ISI Digital Commons (Indian Statistical Institute )
Not a member yet
7571 research outputs found
Sort by
Reinforcement learning-assisted evolutionary algorithm: A survey and research opportunities
No full textEvolutionary algorithms (EA), a class of stochastic search methods based on the principles of natural evolution, have received widespread acclaim for their exceptional performance in various real-world optimization problems. While researchers worldwide have proposed a wide variety of EAs, certain limitations remain, such as slow convergence speed and poor generalization capabilities. Consequently, numerous scholars actively explore improvements to algorithmic structures, operators, search patterns, etc., to enhance their optimization performance. Reinforcement learning (RL) integrated as a component in the EA framework has demonstrated superior performance in recent years. This paper presents a comprehensive survey on integrating reinforcement learning into the evolutionary algorithm, referred to as reinforcement learning-assisted evolutionary algorithm (RL-EA). We begin with the conceptual outlines of reinforcement learning and the evolutionary algorithm. We then provide a taxonomy of RL-EA. Subsequently, we discuss the RL-EA integration method, the RL-assisted strategy adopted by RL-EA, and its applications according to the existing literature. The RL-assisted procedure is divided according to the implemented functions including solution generation, learnable objective function, algorithm/operator/sub-population selection, parameter adaptation, and other strategies. Additionally, different attribute settings of RL in RL-EA are discussed. In the applications of RL-EA section, we also demonstrate the excellent performance of RL-EA on several benchmarks and a range of public datasets to facilitate a quick comparative study. Finally, we analyze potential directions for future research. This survey serves as a rich resource for researchers interested in RL-EA as it overviews the current state-of-the-art and highlights the associated challenges. By leveraging this survey, readers can swiftly gain insights into RL-EA to develop efficient algorithms, thereby fostering further advancements in this emerging field
Secondary transpose of matrix and generalized inverses
No full textIn this paper, several existing results related to secondary transpose are critically reviewed and a result analogous to spectral decomposition theorem is obtained for a real secondary symmetric matrix. Noting that Moore–Penrose inverse with reference to secondary transpose involution, namely s-g inverse, need not always exist, we explore a few necessary sufficient conditions for the existence of such Moore–Penrose inverse. Further, we provide expressions and determinantal formula to compute the same
Self-Supervised Representation Learning for Knee Injury Diagnosis From Magnetic Resonance Data
No full textIn medical image analysis, the cost of acquiring high-quality data and annotation by experts is a barrier in many medical applications. Most of the techniques used are based on a supervised learning framework and require a large amount of annotated data to achieve satisfactory performance. As an alternative, in this article, we propose a self-supervised learning approach for learning the spatial anatomical representations from the frames of magnetic resonance (MR) video clips for the diagnosis of knee medical conditions. The pretext model learns meaningful context-invariant spatial representations. The downstream task in our article is a class-imbalanced multilabel classification. Different experiments show that the features learned by the pretext model provide competitive performance in the downstream task. Moreover, the efficiency and reliability of the proposed pretext model in learning representations of minority classes without applying any strategy toward imbalance in the dataset can be seen from the results. To the best of our knowledge, this work is the first of its kind in showing the effectiveness and reliability of self-supervised learning algorithms in imbalanced multilabel classification tasks on MR scans
Serum miRNA profiling identified miRNAs associated with disease severity in psoriasis
No full textPsoriasis vulgaris is a chronic, autoimmune skin disease involving a complex interplay of epidermal keratinocytes, dermal fibroblast and infiltrating immune cells. Differential expressions of miRNAs are observed in psoriasis and the deregulated miRNAs are sometimes associated with disease severity. This study aims to identify miRNAs altered in the serum of psoriasis patients that are associated with the Psoriasis Area and Severity Index (PASI). In order to assess miRNA levels in the serum of psoriasis patients, we selected 24 differentially expressed miRNAs in the psoriatic skin are possibly derived from the skin and immune cells, as well as five miRNAs that are enriched in other tissues. We identified 16 miRNAs that exhibited significantly (p \u3c 0.05) altered levels in the serum of psoriasis patients compared to healthy individuals. Among these, 13 miRNAs showed similar expression pattern in the serum of psoriasis patients as also observed in the psoriatic skin tissues. Ten miRNAs showed an accuracy of greater than 75% in classifying the psoriasis patients from healthy individuals. Further analysis of differential miRNA levels between the low PASI group and the high PASI group identified three miRNAs (miR-147b, miR-3614-5p, and miR-125a-5p) with significantly altered levels between the low severity and the high severity psoriasis patients. Our systematic investigation of skin and immune cell-derived miRNAs in the serum of psoriasis patients revealed alteration in miRNA levels to be associated with disease severity, which may help in monitoring the disease progression and therapeutic response
Simultaneous non-vanishing of central values of GL(2)×GL(3) and GL(3) L-functions
No full textWe study simultaneous non-vanishing of [Formula presented] and [Formula presented], when Ϝ runs over an orthogonal basis of the space of Hecke-Maass cusp forms for SL(3,Z) and g is a fixed SL(2,Z) holomorphic Hecke cusp form of weight k≡0(mod4)
Some results on the compactified Jacobian of a nodal curve
No full textLet Y be an integral nodal curve. We show that the connected component of the moduli space of torsion free sheaves of rank 1 on the compactified Jacobian J¯(Y) of Y, which contains Pic0J¯(Y), is isomorphic to J¯(Y) under the map induced by the Abel–Jacobi embedding of Y in J¯(Y). We determine the Chern classes (in Chow group) of the Picard bundles on the desingularisation of the compactified Jacobian over a nodal curve Y. We study the relation between the singular cohomology of J¯(Y), J~(Y) and J(X) and use it to determine the singular cohomology of the compactified Jacobian of an integral nodal curve. We prove that the compactified Jacobian of an integral nodal curve with k nodes is homeomorphic to the product of the Jacobian of the normalisation X0 and k rational nodal curves of arithmetic genus 1
Spinning black hole in a fluid
No full textIn this paper, we propose a new analog gravity example - a spinning (or Kerr) black hole in an extended fluid model; the latter was derived in an earlier work [A. K. Mitra and S. Ghosh, Divergence anomaly and Schwinger terms: Towards a consistent theory of anomalous classical fluids, Phys. Rev. D 106, L041702 (2022).PRVDAQ2470-001010.1103/PhysRevD.106.L041702] by two of the present authors. The fluid model receives Berry curvature contributions and applies to electron dynamics in condensed matter lattice systems in the hydrodynamic limit. We construct the acoustic metric for sonic fluctuations that obey a structurally relativistic wave equation in an effective curved background. In a novel approach of dimensional analysis, we have derived explicit expressions for effective mass and angular momentum per unit mass in the acoustic metric (in terms of fluid parameters), to identify with corresponding parameters of the Kerr metric. The spin is a manifestation of the Berry curvature-induced effective noncommutative structure in the fluid. Finally we put the Kerr black hole analogy in a robust setting by revealing explicitly the presence of horizon and ergoregion for a specific background fluid velocity profile. We also show that near horizon behavior of the phase-space trajectory of a probe particle agrees with Kerr black hole analogy. In a fluid dynamics perspective, the presence of a horizon signifies the wave blocking phenomenon
Strain-controlled charge and spin current rectifications in spin-orbit coupled graphene nano-ribbon: A new proposition
No full textIn this work, we investigate the possibilities of performing charge and spin current rectifications using graphene nano-ribbon in the presence of Rashba spin-orbit (SO) interaction. More specifically, we explore the specific role of mechanical strain on these two different types of current rectifications. The system is simulated by a tight-binding framework, where all the results are worked out based on the standard Green’s function formalism. In order to have current rectification, an asymmetry is required, which is incorporated through uncorrelated disorder among the constituent lattice points. From our extensive numerical analysis, we find that reasonably large charge and spin current rectifications can be obtained under strained conditions, and all the physical pictures are valid for a broad range of tight-binding parameters. The rectification properties are studied mostly for zigzag graphene nano-ribbons; however, an armchair ribbon is also taken into account for a clear comparison. Our work may provide a new direction of getting strain-controlled current rectifications in similar kinds of other physical systems as well
Strategy-proof interval-social choice correspondences over extended single-peaked domains
No full textWe consider a social choice model where voters have single-peaked preferences over a finite and ordered set of alternatives that are aggregated to produce contiguous sets or intervals of a fixed cardinality. This is applicable in situations where the alternatives can be arranged in a line (e.g. plots of land) and a contiguous subset of these is required (e.g. a hospital or a school). We define interval-social choice correspondences (I-SCCs) on profiles of single-peaked preferences which select intervals. We extend single-peaked preferences to intervals using responsiveness. We show that generalized median-interval (GMI) rules are the only strategy-proof, anonymous and interval efficient I-SCCs. These rules are interval versions of generalized median voter rules which consist of the median, min and max rules. We show that responsiveness over intervals is necessary for the strategy-proofness of the GMI rule if preferences over alternatives are single-peaked
The Agrarian Economy: Perceptions Versus Reality
No full textUsing evidence from official statistics as well as village studies on incomes from farming across regions, crops, socio-economic classes, and castes, first, I argue that the large majority of cultivators today receive abysmally low incomes and face high variability in incomes. Rising distress can also be inferred from the fact that costs of production have risen faster than government-announced minimum support prices for most crops in most regions of India. Second, I examine the myth of farming in India being highly subsidised, resulting in the overproduction of rice and wheat. I show that subsidies in India are low in comparison to the rest of the world, the European Union and North America, in particular. I also examine projections for different food crops to examine the question of ‘overproduction’. The third issue examined in this article is the invisibility of women workers in agriculture. I argue that women are playing an increasingly significant role in the agricultural economy, a role that official statistics are unable to capture