Detection and Classification of Whale Acoustic Signals

This dissertation focuses on two vital challenges in relation to whale acoustic signals: detection and classification.

In detection, we evaluated the influence of the uncertain ocean environment on the spectrogram-based detector, and derived the likelihood ratio of the proposed Short Time Fourier Transform detector. Experimental results showed that the proposed detector outperforms detectors based on the spectrogram. The proposed detector is more sensitive to environmental changes because it includes phase information.

In classification, our focus is on finding a robust and sparse representation of whale vocalizations. Because whale vocalizations can be modeled as polynomial phase signals, we can represent the whale calls by their polynomial phase coefficients. In this dissertation, we used the Weyl transform to capture chirp rate information, and used a two dimensional feature set to represent whale vocalizations globally. Experimental results showed that our Weyl feature set outperforms chirplet coefficients and MFCC (Mel Frequency Cepstral Coefficients) when applied to our collected data.

Since whale vocalizations can be represented by polynomial phase coefficients, it is plausible that the signals lie on a manifold parameterized by these coefficients. We also studied the intrinsic structure of high dimensional whale data by exploiting its geometry. Experimental results showed that nonlinear mappings such as Laplacian Eigenmap and ISOMAP outperform linear mappings such as PCA and MDS, suggesting that the whale acoustic data is nonlinear.

We also explored deep learning algorithms on whale acoustic data. We built each layer as convolutions with either a PCA filter bank (PCANet) or a DCT filter bank (DCTNet). With the DCT filter bank, each layer has different a time-frequency scale representation, and from this, one can extract different physical information. Experimental results showed that our PCANet and DCTNet achieve high classification rate on the whale vocalization data set. The word error rate of the DCTNet feature is similar to the MFSC in speech recognition tasks, suggesting that the convolutional network is able to reveal acoustic content of speech signals.

They’re There, Now What?: The Identities, Behaviors, and Perceptions of Black Judges

Prior to the Civil Rights Movement, fewer than 50 Black judges had been elected or appointed to the judiciary. As of August 2015, there are over 1,000 Black state and federal judges. As the number of black judges has increased, one question arises: have American courts been altered purely by this substantial increase? One expectation—and, at times, a prediction—behind the increased descriptive representation of Black judges is that their mere presence would alter the judiciary. It was supposed that these judges would substantively represent Black interests in the decisions they made. In other words, it was suspected, and predicted, that Blacks in the judiciary would enhance equality and justice by being aware of, responsive to, and advocating for African Americans. This theory about the likely role of Black judges derives from theoretical work on political representation and racial group consciousness, and empirical studies of Black elite behavior in other political institutions.

Despite such predictions, there is no corresponding scholarly consensus regarding whether Black judges possess a racial group consciousness and have racially distinctive judicial behavior. Therefore, the theory undergirding the demand for increased diversification, as a means to transform the judiciary, remains unsubstantiated. This is precisely where this project, “They’re There, Now What?: The Identities, Behavior, and Perceptions of Black Judges,” seeks to intervene in and explore, if not settle, the matter of whether black judges possess a racial group consciousness and exhibit racially-distinctive judicial behavior. It addresses a set of interrelated questions relevant to understanding whether we can view Black judges as representatives in ways that are similar to how we view other Black political officials. I examine these questions using a multi-method approach. For my analyses, I draw on diverse materials: the published biographies of every Black judge appointed to the federal bench, a survey experiment with a nationally-representative adult sample, and semi-structured interviews with 30 Black judges.

This research, which engages with scholarship on representation, group consciousness, judicial behavior, and candidate perceptions, offers new insights into the lives, perceptions, and behavior of Black judges, as well as the manifestations of Black substantive representation in the judiciary. My dissertation argues that, despite the general reluctance to use the term “representation” when referring to judges, we can consider Black judges as representatives. Black judges behave as substantive representatives by (1) sharing and understanding the experience, history, and perspectives of Black Americans, (2) challenging language, persons, policies, and laws they feel negatively affect, or violate the rights and liberties of, African Americans, (3) respecting African American litigants, and (4) ensuring the rights of African Americans are protected and the needs of black Americans are being met.

Only through research that considers the perspectives, identities, perceptions, and behavior of Black judges will we arrive at a more comprehensive understanding of the importance of racial diversity in the courts. As this project finds, a link between descriptive representation and substantive representation can, and frequently does exist within the judicial context. Such a link is significant given that Blacks’ liberty and justice through the American legal system continues to be subject to those who exercise judicial power. This dissertation has implications for the discourse surrounding the need for increased descriptive and substantive representation of Blacks in the judiciary, and the factors that affect representation in the justice system.