Spring

Neural Architecture Search (NAS) is a method fpr automating the process of designing neural network architectures by using algorithms such as evolutionary algorithms or reinforcement learning to search through a predefined space of architectures. NAS is used by a couple of open-source AutoML tools such as AutoKeras, AutoPytorch and AutoGluon in order to find a DNN model with best performance for a given task.

The first step in nearly all natural language processing (NLP) applications is applying preprocessing operations [1] to the text. Preprocessing operations include tokenization (segmenting the text into tokens), sentence splitting (dividing the text into sentences), normalization (converting the text into a canonical form), and the like. In this project, you will develop and implement algorithms for preprocessing of Turkish text using deep learning approaches. First, a literature review will be conducted and similar systems for English will be analyzed (e.g. UDPipe [2], Stanza [3]).

    In this project, you will design and implement a web service to publish senior project offers from faculty to students and disseminate the project results to a wider audience.

    Roadmap:

    - Requirements collection (talk with your fellow students and faculty)
    - Survey of available technologies
    - Designing the service according to requirements and available technologies
    - Implementation and testing

    Other notes:

Bevy is an open-source ECS based game engine written in Rust programming language.  https://bevyengine.org/    https://github.com/bevyengine/bevy
ECS (Entity-Component-System) is a software pattern which is generally used in computer games.
Bevy is a novel project which is not as mature as other game engines like Unity, Godot, Unreal etc. Therefore, currently it lacks a built in GUI based profiler to collect performance metrics when games are run on it.

Locality Sensitivity Hashing (LSH) is a powerful approach to similarity (or distance) estimation, which exploits a family of randomized hash functions to map the similar data instances to the same buckets with a higher probability than the dissimilar ones. In other words, LSH algorithms produce comparable hash values for similar values unlike the cryptographic hash algorithms that produces entirely different outputs for similar values.