Machine learning successfully predicts fly behavior from recorded neural activity
- Publications
- April 29, 2022
Though seemingly insubstantial in size, flying insects such as mosquitoes and fruit flies significantly impact humans. Specific species of fruit flies are responsible for widespread crop damage and spoilage whereas certain female mosquitoes harbor pathogens in their salivary glands, transmitting them to humans when biting (or blood feeding). Humans have historically addressed the issue with high-volume insecticide spraying throughout at-risk …
View Post
Chemicals naturally on human skin interfere with mosquito behavior
- Publications
- February 14, 2018
Receiver operating characteristic (ROC) analysis is a method for evaluating successful discrimination. The machine learning model (SVM) predicts chemicals that were not in the training data. Predictions for these new chemicals are then compared to the ground truth. Success is defined by high positive (sensitivity) and low false-positive (1 − specificity) rates. Subsequently, an ROC plot shows the relationship between these …
View Post
Machine learning advances effort to crack the perceptual code
- Publications
- February 14, 2018
Something mysterious and remarkable resides in the sense of smell, most famously captured by the novelist Marcel Proust in The Remembrance of Things Past. In the years following its original publication, scientists were examining the mechanisms that gave rise to scent profiles such as lemon and rose. It was already known that visual and auditory perceptual qualities were not complete …
View Post
