Source author record

Mukul Kumar

Mukul Kumar appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

cond-mat.mtrl-sci Machine Learning Computation and Language cond-mat.mes-hall Information Retrieval

Catalog footprint

What is connected

5works

5topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2020arXiv

A Weighted Mutual k-Nearest Neighbour for Classification Mining

kNN is a very effective Instance based learning method, and it is easy to implement. Due to heterogeneous nature of data, noises from different possible sources are also widespread in nature especially in case of large-scale databases. For noise elimination and effect of pseudo neighbours, in this paper, we propose a new learning algorithm which performs the task of anomaly detection and removal of pseudo neighbours from the dataset so as to provide comparative better results. This algorithm also tries to minimize effect of those neighbours which are distant. A concept of certainty measure is also introduced for experimental results. The advantage of using concept of mutual neighbours and distance-weighted voting is that, dataset will be refined after removal of anomaly and weightage concept compels to take into account more consideration of those neighbours, which are closer. Consequently, finally the performance of proposed algorithm is calculated.

preprint2020arXiv

Nanocrystalline grain boundary engineering: Increasing $Σ$3 boundary fraction in pure Ni with thermomechanical treatments

Grain boundary networks should play a dominant role in determining the mechanical properties of nanocrystalline metals. However, these networks are difficult to characterize and their response to deformation is incompletely understood. In this work, we study the grain boundary network of nanocrystalline Ni and explore whether it can be modified by plastic deformation. Mechanical cycling at room temperature did not lead to structural evolution, but elevated temperature cycling did alter the grain boundary network. In addition to mechanically-driven grain growth, mechanical cycling at 100 $°$C led to a 48% increase in $Σ$3 boundaries, determined with transmission Kikuchi diffraction. The extent of boundary modification was a function of the number of applied loading cycles and the testing temperature, with more cycles at higher temperatures leading to more special grain boundaries. The results presented here suggest a path to grain boundary engineering in nanocrystalline materials.

preprint2019arXiv

Shareable Representations for Search Query Understanding

Understanding search queries is critical for shopping search engines to deliver a satisfying customer experience. Popular shopping search engines receive billions of unique queries yearly, each of which can depict any of hundreds of user preferences or intents. In order to get the right results to customers it must be known queries like "inexpensive prom dresses" are intended to not only surface results of a certain product type but also products with a low price. Referred to as query intents, examples also include preferences for author, brand, age group, or simply a need for customer service. Recent works such as BERT have demonstrated the success of a large transformer encoder architecture with language model pre-training on a variety of NLP tasks. We adapt such an architecture to learn intents for search queries and describe methods to account for the noisiness and sparseness of search query data. We also describe cost effective ways of hosting transformer encoder models in context with low latency requirements. With the right domain-specific training we can build a shareable deep learning model whose internal representation can be reused for a variety of query understanding tasks including query intent identification. Model sharing allows for fewer large models needed to be served at inference time and provides a platform to quickly build and roll out new search query classifiers.

preprint2015arXiv

Grain boundary character distributions in nanocrystalline metals produced by different processing routes

Nanocrystalline materials are defined by their fine grain size, but details of the grain boundary character distribution should also be important. Grain boundary character distributions are reported for ball milled, sputter deposited, and electrodeposited Ni and Ni-based alloys, all with average grain sizes of ~20 nm, to study the influence of processing route. The two deposited materials had nearly identical grain boundary character distributions, both marked by a Σ3 length percentage of 23-25%. In contrast, the ball milled material had only 3% Σ3-type grain boundaries and a large fraction of low angle boundaries (16%), with the remainder being predominantly random high angle (73%). These grain boundary character measurements are connected to the physical events that control their respective processing routes. Consequences for material properties are also discussed with a focus on nanocrystalline corrosion. As a whole, the results presented here show that grain boundary character distribution, which has often been overlooked in nanocrystalline metals, can vary significantly and influence material properties in profound ways.

preprint2013arXiv

Anisotropy of Interfacial Energy in Five Dimensions

Anisotropy of interfacial energy is the principal driving force for material microstructure evolution yet its origins remain uncertain and a quantitative description lacking. We present and justify a concise hypothesis on the topography and topology of the functional space of grain boundary energies and, based on this hypothesis, construct a closed-form function that quantitatively describes energy variations in the entire 5-space of macroscopic parameters defining grain boundary geometry. The new function is found to be universal for the crystallography class of face-centered cubic metals.