Prediction of chip failures at early stages of production flow
Evaluation of DL keypoint-matching approaches for structure from motion
Pulmonary Embolism Identification
Predicting formation of dry areas in DSW evaporation ponds
Build a model which predicts failures of manufactured chips based on indicators coming from the different stages of production.
Train deep learning models to detect, describe and match keypoints and fine-tune these models to specific domains of chronic wound images.
Build an algorithm aimed at the detection and classification of PE cases based on a Kaggle freely available data set of chest CTPA images.
Predict the formation of dry areas in both the Salt and Carnallite Ponds.
Port mapping using behavioral vessel data
Automatic representation of molecules as complicated features
Use vessel behavioral data in order to map port-related areas of interest.
Build a high performance classifier that predicts the diagnosis of the physician using the information collected in the visit.
Develop a matching algorithm that will recommend a match between applications and domains.
Develop an property prediction algorithm based on advanced molecular features.
Speeding up Transformer-based NLP models
Train NLP models based on BERT, ELECTRA, ROBERTA and other models using a GPU, and then experiment with various methods to reduce their complexity and run times on a CPU.
Cloaking Score for Taboola campaigns
Create ML model that is able to score each new live campaigns in Taboola with cloaking score.