This Article challenges the common wisdom about the desirability of celebrated trademarks. Contrary to the traditional view, it argues that mega-brands are neither economic evils nor is their function limited to imparting information regarding the physical product they flaunt. The Article also rejects the view that famous marks persuade consumers (often referred to as “Snobs”) to “irrationally” pay more for the same physical product they could have purchased for less. Rather, it argues that in purchasing a branded good the consumer is actually purchasing three tied products in one package: a physical product, information about the physical product, and an intangible product such as fame, prestige, peace of mind or a pleasant feeling. Contrary to prior literature, this Article argues that the intangible product benefits both producers and consumers. It explores the demand for the intangible product, its impact on pricing, welfare and consumers’ and producers’ strategies. It argues that under certain conditions one may witness the anomaly of an increase in both price and output, but that such observation does not mandate the conclusion that consumers are facing an up-sloping demand curve as discussed in the literature of conspicuous goods. Instead, the Article proposes that this phenomenon may occur in the traditional down-sloping demand curves and that it is not limited to goods with conspicuous properties. The Article has normative and descriptive implications with regard to three distinct bodies of law: price discrimination, trademark anti-dilution and trade-name law. A direct result is that mega-brands neither confer a monopoly nor foster price discrimination. On the contrary, they enhance competition in both the physical and intangible spheres. The Article also offers a new rational basis for one of the most nebulous doctrines in trademark law: anti-dilution. Anti-dilution law provides special protection to famous marks which is not available to regular ones, and has been unanimously enthroned as one which protects only producers but is injurious to consumers. Conversely, this Article argues that anti-dilution law inures to the benefit of both consumers and producers. It attempts to clear the constitutional concerns that have been raised with regard to the doctrine and explains the fame requirement. The article concludes that Snobs are rational and that there are sound economic justifications for the law’s unique protection of famous marks.

This dissertation examines how the sequence and the chemical environment of intrinsically disordered protein regions (IDRs) affect their structure. Unlike structured proteins, IDRs do not adopt a singular structure. Instead, they exist in a dynamic conformational ensemble. The conformations that make up this ensemble are shaped by a range of molecular interactions, both with the environment and within the IDR itself. The absence of a stable three-dimensional structure, along with their high level of exposure to the solvent environment, makes IDR exceptionally adaptable to changes in their physical and chemical surroundings. My research employs computational methods to investigate how IDRs respond to different chemical environments and physical constraints. I will discuss how we quantify the contribution of IDR structural preferences to their entropic forces and describe how we leverage IDR sensitivity to design biosensors.

Legal rights may erode as a result of past, uncontested, breach. In light of ongoing violations, the rightholder's enforcement may result in the loss of the entitlement. The doctrines, of course, of performance in contract law and adverse possession in property law are prominent examples of this widespread erosion phenomenon. In analyzing the effects of such laws, the Article confronts two conflicting intuitions. On the one hand, the "license" to continue breach prospectively encourages opportunism. On the other hand, the risk of erosion may reinforce the rightholder's motivation to take anti-erosion measures, bolstering the credibility of the threat to enforce, thus better preserving the entitlement. The article proves that these two effects of erosion rules always balance out. The same amount of value will be extracted from the right holder irrespective of the law's erosion doctrine. The article also demonstrates the limits of this "irrelevance" claim and the factors that may lead to its collapse. It applies the analysis to offer new perspectives on various prominent legal rules.

Legal rights may erode as a result of past, uncontested, breach. In light of ongoing violations, the rightholder's enforcement may result in the loss of the entitlement. The doctrines, of course, of performance in contract law and adverse possession in property law are prominent examples of this widespread erosion phenomenon. In analyzing the effects of such laws, the Article confronts two conflicting intuitions. On the one hand, the "license" to continue breach prospectively encourages opportunism. On the other hand, the risk of erosion may reinforce the rightholder's motivation to take anti-erosion measures, bolstering the credibility of the threat to enforce, thus better preserving the entitlement. The article proves that these two effects of erosion rules always balance out. The same amount of value will be extracted from the right holder irrespective of the law's erosion doctrine. The article also demonstrates the limits of this "irrelevance" claim and the factors that may lead to its collapse. It applies the analysis to offer new perspectives on various prominent legal rules.

Anomaly detection is a data partitioning algorithm which separates outliers from normative data points. An unsupervised learning approach to this problem does not assume any prior information. Anomaly detection is a primary data analysis task with diverse applications and has been studied under many models and assumptions.Spectral methods such as spectral clustering have been widely used to solve the clustering problem. These methods make use of the leading K eigenvectors of the graph Laplacian matrix to detect K clusters, if the graph has a clear community structure. In a setting where the data consists of unbalanced clusters, as in anomaly detection, the spectral properties are determined by a dominating component. In such cases, traditional graph clustering methods fail, while the spectral embedding norm was found to overcome this challenge. This thesis generalizes the spectral embedding norm definition, formerly used, by introducing the capability of a weighted norm. With just one simple natural condition: allowing limited connectivity between the clusters and the background, we prove that this quantity can be used to detect the clusters of interest. Experiments on both synthetic data sets and real-world defect detection images demonstrate the effectiveness of the algorithm and its performance was found to be stable, with respect to parameter choices.

Intrinsically disordered protein regions (IDRs) play a variety of essential roles in regulating cellular function. Instead of folding into a single stable structure, an IDR exists in an ensemble of interconverting conformations biased by local and long-range intramolecular interactions. The absence of a fixed 3D structure combined with high solvent accessibility makes IDRs sensitive to changes in their physical-chemical surroundings. This dissertation describes investigations related to the ability of IDRs to sense and respond to changes in their surrounding solution conditions, and proposes a new paradigm of IDRs as sensors of cellular physicochemistry. Chapter 1 introduces the concept of IDRs as physicochemical sensors. Chapter 2 describes a project in which I developed a pipeline to allow medium-throughput FRET assays of IDR global dimensions in a wide variety of solution conditions, as well as benchmarking and analytical tools to allow meaningful comparison of the behavior of diverse IDRs, and used this system to characterize the end-to-end distance and solution sensitivity of several naturally-occurring IDRs. Chapter 3 describes a project in which my colleague and I performed parallel assays on naturally-occurring and synthetic IDRs in vitro and in live cells to test whether structural biases and sensitivity seen in vitro would be preserved when the same IDRs were tested in live cells. Chapter 4 describes a bioengineering project in which we used the systems described in Chapters 2 and 3 to contribute to the design and testing of a novel IDR-based biosensor that can monitor osmotic stress in live cells. Chapter 5 concludes the dissertation and suggests directions for further research.