Hypothesis-Driven Deep Learning for Out of Distribution Detection

Predictions of opaque black-box systems are frequently deployed in high-stakes applications such as healthcare. For such applications, it is crucial to assess how models handle samples beyond the domain of training data. While several metrics and tests exist to detect out-of-distribution (OoD) data from in-distribution (InD) data to a deep neural network (DNN), their performance varies significantly across datasets, models, and tasks, which limits their practical use. In this paper, we propose a hypothesis-driven approach to quantify whether a new sample is InD or OoD. Given a trained DNN and some input, we first feed the input through the DNN and compute an ensemble of OoD metrics, which we term latent responses. We then formulate the OoD detection problem as a hypothesis test between latent responses of different groups, and use permutation-based resampling to infer the significance of the observed latent responses under a null hypothesis. We adapt our method to detect an unseen sample of bacteria to a trained deep learning model, and show that it reveals interpretable differences between InD and OoD latent responses. Our work has implications for systematic novelty detection and informed decision-making from classifiers trained on a subset of labels.

Gaze-Net: Appearance-Based Gaze Estimation using Capsule Networks

Recent studies on appearance based gaze estimation indicate the ability of Neural Networks to decode gaze information from facial images encompassing pose information. In this paper, we propose Gaze-Net: A capsule network capable of decoding, representing, and estimating gaze information from ocular region images. We evaluate our proposed system using two publicly available datasets, MPIIGaze (200,000+ images in the wild) and Columbia Gaze (5000+ images of users with 21 gaze directions observed at 5 camera angles/positions). Our model achieves a Mean Absolute Error (MAE) of 2.84$^\circ$ for Combined angle error estimate within dataset for MPI-IGaze dataset. Further, model achieves a MAE of 10.04$^\circ$ for across dataset gaze estimation error for Columbia gaze dataset. Through transfer learning, the error is reduced to 5.9$^\circ$. The results show this approach is promising with implications towards using commodity webcams to develop low-cost multi-user gaze tracking systems.

Electroencephalogram (EEG) for Delineating Objective Measure of Autism Spectrum Disorder (ASD) (Extended Version)

Autism Spectrum Disorder (ASD) is a developmental disorder that often impairs a child's normal development of the brain. According to CDC, it is estimated that 1 in 6 children in the US suffer from development disorders, and 1 in 68 children in the US suffer from ASD. This condition has a negative impact on a person's ability to hear, socialize and communicate. Overall, ASD has a broad range of symptoms and severity; hence the term spectrum is used. One of the main contributors to ASD is known to be genetics. Up to date, no suitable cure for ASD has been found. Early diagnosis is crucial for the long-term treatment of ASD, but this is challenging due to the lack of a proper objective measures. Subjective measures often take more time, resources, and have false positives or false negatives. There is a need for efficient objective measures that can help in diagnosing this disease early as possible with less effort. EEG measures the electric signals of the brain via electrodes placed on various places on the scalp. These signals can be used to study complex neuropsychiatric issues. Studies have shown that EEG has the potential to be used as a biomarker for various neurological conditions including ASD. This chapter will outline the usage of EEG measurement for the classification of ASD using machine learning algorithms.