This work traces the evolving interpretation of property as revealed in key U.S. Supreme Court rulings between 1924 and 1978. The degree of influence of any given case is determined by ranking case citations from the time of ruling to the time of writing. A series of queries was applied to the open-source AI, ChatGPT, as a corroborating source. I find that over this period there exist multiple lines of evolving thought regarding the nature of property and the public purpose. First, economic methods and techniques associated with market valuation are increasingly employed as a method to determine what constitutes property as well as the public interest. Second, despite increasing formal economic input, the court consistently finds "space" within the known economic data in which to exercise their discretion over the meanings of property and the public purpose. Third, there exists a "back and forth" between points one and two; namely, that what is decided based upon court discretion, later becomes a matter of economic fact, which in turn leads to new areas of court discretion.

While there exist a plethora of deep learning tools and frameworks, the fast-growing complexity of the field brings new demands and challenges, such as more flexible network design, speedy computation on distributed setting, and compatibility between different tools. In this paper, we introduce Neural Network Libraries (https://nnabla.org), a deep learning framework designed from engineer's perspective, with emphasis on usability and compatibility as its core design principles. We elaborate on each of our design principles and its merits, and validate our attempts via experiments.

Provided is a sound source separation device that includes a combining unit that combines a first sound source separation signal of a predetermined sound source, the first sound source separation signal being separated from a mixed sound signal by a first sound source separation system, with a second sound source separation signal of the sound source, the second sound source separation signal being separated from the mixed sound signal by a second sound source separation system that differs in separation performance from the first sound source separation system in predetermined units of time, and that outputs a sound source separation signal obtained by the combination.

Modern text-to-speech systems are able to produce natural and high-quality speech, but speech contains factors of variation (e.g. pitch, rhythm, loudness, timbre)\ that text alone cannot contain. In this work we move towards a speech synthesis system that can produce diverse speech renditions of a text by allowing (but not requiring) explicit control over the various factors of variation. We propose a new neural vocoder that offers control of such factors of variation. This is achieved by employing differentiable digital signal processing (DDSP) (previously used only for music rather than speech), which exposes these factors of variation. The results show that the proposed approach can produce natural speech with realistic timbre, and individual factors of variation can be freely controlled.

Recently, predictor-based algorithms emerged as a promising approach for neural architecture search (NAS). For NAS, we typically have to calculate the validation accuracy of a large number of Deep Neural Networks (DNNs), what is computationally complex. Predictor-based NAS algorithms address this problem. They train a proxy model that can infer the validation accuracy of DNNs directly from their network structure. During optimization, the proxy can be used to narrow down the number of architectures for which the true validation accuracy must be computed, what makes predictor-based algorithms sample efficient. Usually, we compute the proxy for all DNNs in the network search space and pick those that maximize the proxy as candidates for optimization. However, that is intractable in practice, because the search spaces are often very large and contain billions of network architectures. The contributions of this paper are threefold: 1) We define a sample efficiency gain to compare different predictor-based NAS algorithms. 2) We conduct experiments on the NASBench-101 dataset and show that the sample efficiency of predictor-based algorithms decreases dramatically if the proxy is only computed for a subset of the search space. 3) We show that if we choose the subset of the search space on which the proxy is evaluated in a smart way, the sample efficiency of the original predictor-based algorithm that has access to the full search space can be regained. This is an important step to make predictor-based NAS algorithms useful, in practice.

Operating deep neural networks (DNNs) on devices with limited resources requires the reduction of their memory as well as computational footprint. Popular reduction methods are network quantization or pruning, which either reduce the word length of the network parameters or remove weights from the network if they are not needed. In this article we discuss a general framework for network reduction which we call `Look-Up Table Quantization` (LUT-Q). For each layer, we learn a value dictionary and an assignment matrix to represent the network weights. We propose a special solver which combines gradient descent and a one-step k-means update to learn both the value dictionaries and assignment matrices iteratively. This method is very flexible: by constraining the value dictionary, many different reduction problems such as non-uniform network quantization, training of multiplierless networks, network pruning or simultaneous quantization and pruning can be implemented without changing the solver. This flexibility of the LUT-Q method allows us to use the same method to train networks for different hardware capabilities.

This article documents the development and implementation of an undergraduate research methods course incorporating field work. Conducted in collaboration between the UW-Eau Claire Department of Economics and local elected officials and county staff in Buffalo County, Wisconsin undergraduate students were assigned the task of developing and presenting initiating a county-wide economic redevelopment strategy during the Spring term 2018. Course specific learning goals included: Hard skills development (software usage, data acquisition and analysis, benchmarking, and survey design) and soft skills development (presentation of sensitive economic information to lay audiences and teamwork). Initial results suggest that field-based coursework improves student learning, student interest in economics, improve local economic performance, and may help to bridge the "Town versus Gown" divide.

Music source separation with deep neural networks typically relies only on amplitude features. In this paper we show that additional phase features can improve the separation performance. Using the theoretical relationship between STFT phase and amplitude, we conjecture that derivatives of the phase are a good feature representation opposed to the raw phase. We verify this conjecture experimentally and propose a new DNN architecture which combines amplitude and phase. This joint approach achieves a better signal-to distortion ratio on the DSD100 dataset for all instruments compared to a network that uses only amplitude features. Especially, the bass instrument benefits from the phase information.

An electronic device comprising circuitry configured to: obtain information representing a person's knowledge and/or skills; update a user model representing the person's knowledge and/or skills based on the obtained information; and provide user-adaptive information depending on the user model.

Ein mobiles Gerat umfasst einen Sensor und einen Prozessor. Der Sensor ist dazu ausgebildet, drahtlos ein einen Gegenstands-Kennzeichnungscode enthaltendes Identifikationssignal zu empfangen, wobei das Identifikationssignal von einem Low-Rate-Ubertragungsgerat ausgesendet wird, das einem Gegenstand zugeordnet ist, wobei der Sensor das Identifikationssignal aus Richtung des Gegenstands empfangt. Der Prozessor ist dazu ausgebildet, basierend auf dem Gegenstands-Kennzeichnungscode ein Ausgabesignal zu erzeugen.

This paper deals with the separation of music into individual instrument tracks which is known to be a challenging problem. We describe two different deep neural network architectures for this task, a feed-forward and a recurrent one, and show that each of them yield s themselves state-of-the art results on the SiSEC DSD100 dataset. For the recurrent network, we use data augmentation during training and show that even simple separation networks are prone to overfitting if no data augmentation is used. Furthermore, we propose a blending of both neural network systems where we linearly combine their raw outputs and then perform a multi-channel Wiener filter post-processing. This blending scheme yields the best results that have been reported to-date on the SiSEC DSD100 dataset.

This paper deals with the separation of music into individual instrument tracks which is known to be a challenging problem. We describe two different deep neural network architectures for this task, a feed-forward and a recurrent one, and show that each of them yields themselves state-of-the art results on the SiSEC DSD100 dataset. For the recurrent network, we use data augmentation during training and show that even simple separation networks are prone to overfitting if no data augmentation is used. Furthermore, we propose a blending of both neural network systems where we linearly combine their raw outputs and then perform a multi-channel Wiener filter post-processing. This blending scheme yields the best results that have been reported to-date on the SiSEC DSD100 dataset.

An electronic device comprising an input interface adapted to receive an input signal, a command determination unit that is adapted to determine a plurality of possible commands based on the input signal, and an output interface, which is adapted to receive a plurality of output information provide corresponding effects associated with each of the plurality of specific possible commands.