Spring

Develop a runtime verification tool in Go or Rust

Runtime verification is a computing system analysis and execution approach based on extracting information from a running system and using it to detect and possibly react to observed behaviors satisfying or violating certain properties. Reelay is a header-only C++ library for system-level verification and testing of real-time systems. Reelay implements state-of-the-art runtime verification techniques to construct runtime monitors that check temporal behaviors of the system against system-level requirements.

Opinion dynamics in Online Social Networking Systems

Social networking platforms are widely used to influence opinions across all kinds of domains. The high volume of short messages along with the characteristics of posts makes identifying the opinions and the dynamics of opinions rather challenging.

In this project, you will develop a program to identify and track opinions on online social networks.

This project can be formulated as Cmpe491 or Cmpe492 depending on interest.

Bladder Segmentation to Classify the Level of Tumor Invasions According to VI-RADS

This is an interdisciplinary project that aims to classify the degree to which of bladder lesions invade the bladder walls, which is critical to assessing the patient's treatment.
The project is inpired by VI-RADS (a standardization of MRI acquisition and interpretation that yields a score in range [1, 5]) that has been shown to be effective in medical studies.

In this project, deep learning methods will be explored as a classifier of the scale of bladder lesions. It includes elicitation sessions with radiologists.

Application of Federated Learning on Wearable Data (INF491 or INF492)

FL is a new methodology providing privacy in learning by sharing only model parameters rather than the data directly with the server. In this project, we aim to apply well-known FL techniques to wearable data collected from university students and workers and compare them with traditional ML techniques. The dataset includes activity, sleep, and stress-related information.

A Comparative Study of Simulators for Sim-To-Real Deep Reinforcement Learning

Note that this project will be co-supervised by Suzan Ece Ada, Phd student in Colors Lab.

Object - Robot Interaction Type Detection

Humans can interact with objects differently by using only one hand. The grippers on the robot can be used for different actions too. Via using the contact points on the objects and detecting the movements at these points. The actions should be recognized.

Using large-scale language models to learn robust robotic representations

Note that this project will be co-supervised by Alara Dirik from Colors Lab.

Extracting Markers from physiological and survey data collected from wearable devices (INF491 or INF492)

Wearable devices can capture multimodal data corresponding to a person’s activity, stress, sleep information to measure and improve health and well-being. Besides device measurement, there are also survey data as another information gathering methodology.

Extracting Markers from physiological and survey data collected from wearable devices

Wearable devices can capture multimodal data corresponding to a person’s activity, stress, sleep information to measure and improve health and well-being. Besides device measurement, there are also survey data as another information gathering methodology.

Human Robot Interaction Application with Baxter

Baxter is an upper-body humanoid robot platform with two arms and a face:

https://en.wikipedia.org/wiki/Baxter_(robot)

The aim of this project is to write an application to create interactive and appealing scenarios for Baxter and user communication.

Arama formu

Main Menu

Spring

Develop a runtime verification tool in Go or Rust

Opinion dynamics in Online Social Networking Systems

Bladder Segmentation to Classify the Level of Tumor Invasions According to VI-RADS

Application of Federated Learning on Wearable Data (INF491 or INF492)

A Comparative Study of Simulators for Sim-To-Real Deep Reinforcement Learning

Object - Robot Interaction Type Detection

Using large-scale language models to learn robust robotic representations

Extracting Markers from physiological and survey data collected from wearable devices (INF491 or INF492)

Extracting Markers from physiological and survey data collected from wearable devices

Human Robot Interaction Application with Baxter

Sayfalar

Bize Ulaşın

Bizi takip edin

Arama formu

Main Menu

Spring

Sayfalar

Bize Ulaşın

Bizi takip edin

You are here: