Digital Commons Network^™

Overview Of Open Source Codes To Assess Environmental Effects Of Ocean Wave Farms (Extended Abstract), Chris Chartrand, Kelley Ruehl, Jesse Roberts, Aaron Porter, Cameron Mcnatt

Ocean Waves Workshop

The United States has a theoretical ocean wave energy resource potential of 1,594–2,640 TWh/year, enough to power between 143.5 and 237.6 million homes/year and contribute substantially to the United States’ energy portfolio [1]. However, wave energy converters (WECs) are currently in the early stages of research and development at low technology readiness levels. Open ocean deployment data is from demonstration-scale projects, not from utility-scale deployments. As a result, researchers, developers, and regulators rely heavily on numerical models to understand the environmental effects of wave farms.

Preliminary numerical studies have demonstrated that small-scale deployments of ~10 WECs or less have little …

Session 2 Presentation - A New Generation Of Spectral Wave Models, Don Resio, Dorukan Ardag, C. Reid Nichols

Ocean Waves Workshop

Over the last decade, considerable effort has been spent on the evaluation of operational wave models such as the WAM, WAVEWATCH III, and SWAN thirdgeneration (3G) model. Most recent effort has focused on a “wholistic” approach to model evaluation, rather than on related comparisons directly to physics-based processes. In such an approach, model coefficients are tuned to maximize the global fit to integrated wave parameters such as wave height, mean/peak period, mean direction and directional spreading [1]. However, these coefficients do not ensure a universal “best-fit” to observation. Also, comparisons to spectra are omitted from the model evaluations since they …

Machine Learning For Hardware Simulation, N/A

Defensive Publications Series

Generally, the present disclosure is directed to improving hardware simulation through machine learning. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict hardware test results based on data extracted from hardware.

Contextual Advertising Based On Content Recognition In A Video, Jason Mayes

Defensive Publications Series

Generally, the present disclosure is directed to providing relevant advertisements based on the visual content of a video. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to determine a relevant advertisement and/or a relevant time for the relevant advertisement based on image data taken from a video.

Identifying And Categorizing Utility Power Anomalies, N/A

Defensive Publications Series

Generally, the present disclosure is directed to predicting utility power anomalies such as, for example, outage events. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict utility power anomalies based on historical and/or current utility power data.

Recognizing Semantic Formatting Information In A Document, N/A

Defensive Publications Series

Generally, the present disclosure is directed to recognizing semantic intent from formatting information. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict the intent of an author for a section (e.g., a word, sentence, line, column and/or paragraph) of a document based on the formatting information of that section of the document.

Suggesting Deletion Of Blurry Photos, Charles Lai, Nicolas Miranda

Defensive Publications Series

Generally, the present disclosure is directed to identifying and suggesting deletion of blurry photos. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict a blurriness characteristic of an image based on image data. For example, the blurriness characteristic can describe a percentage of the image that is blurry.

Learning Better Font Slicing Strategies From Data, Quan Wang, Yiran Mao

Defensive Publications Series

Generally, the present disclosure is directed to serving font files in topical subsets. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict topic labels of characters in a font based on a corpus of characters or glyphs.

Session 1 Discussion Notes, Ocean Waves Workshop 2017

Ocean Waves Workshop

No abstract provided.

Using Imagery Of Vegetation And Rooftops To Predict Solar Roof Candidacy, Charles Armstrong

Defensive Publications Series

Generally, the present disclosure is directed to predicting whether a property can be conducive to solar energy system installation with minimal adjustment to peripheral vegetation. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict whether a property can be conducive to solar energy installation with minimal adjustment to peripheral vegetation based on imagery and/or publicly available or user-submitted information about the property or area surrounding the property.

Using Imagery To Identify Abandoned Property In Public Spaces, Charles Armstrong

Defensive Publications Series

Generally, the present disclosure is directed to identifying private property that has been illegally dumped or stored for a prolonged period without use. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict whether an item has been abandoned in a public right-of-way based on imagery of the item and record of public right-of-way.

System Volume Compensating For Environmental Noise, Nick Felker

Defensive Publications Series

Generally, the present disclosure is directed to an audio system for compensating for ambient environmental noise. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict a comfortable volume level based on an intensity of ambient noise.

Using Music To Affect Mood Based On Sentiment Analysis, Nick Felker

Defensive Publications Series

Generally, the present disclosure is directed to influencing the mood of a person listening to music while operating a system or device. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to select music to play to influence a user’s mood based on user input into a system or device.

Planning Group Meals Based On Preferences Of Attendees, Nick Felker

Defensive Publications Series

Generally, the present disclosure is directed to determining an optimal place and time for a meeting based on the preferences of the people attending the meeting. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict an optimal restaurant and time for a group meal based on personal preferences and/or time availability of members of the group.

Font Optimization Based On User Reading Habits, Nick Felker

Defensive Publications Series

Generally, the present disclosure is directed to improving accessibility to electronic documents by adjusting font characteristics based on user behavior. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict an optimal font to display an electronic document in based on user interaction with the electronic document.

Detecting Interesting Events In A Home Security Camera System, N/A

Defensive Publications Series

Generally, the present disclosure is directed to a system for predicting whether the subject of a camera needs to be recorded and/or transmitted. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict whether the view of the camera contains a noteworthy change in semantic meaning based on labels describing the semantic meaning of the view.

Machine-Learning For Optimization Of Software Parameters, Etienne J. Membrives

Defensive Publications Series

Generally, the present disclosure is directed to optimizing tuning parameters in a computing system and/or software application using machine learning. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict an optimal value for tuning parameters based on metrics provided by a developer. As examples, such metrics may be related to an amount of user engagement, latency associated with the application, or efficiency of executing the software application.

Providing Relevant Advertisements Based On Item-Specific Purchase History, Nick Felker

Defensive Publications Series

Generally, the present disclosure is directed to providing relevant advertisements to a shopper based on purchase history. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict how receptive a shopper will be to advertisements for a similar item to an item recently purchased based on purchase history patterns for the item.

Machine-Learned Caching Of Datasets, Etienne J. Membrives

Defensive Publications Series

Generally, the present disclosure is directed to creating and/or modifying a pre-cache for a client device connected to a remote server containing a dataset. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict the likelihood a particular piece of data will be used (e.g. opened, edited, saved, etc.) within a time frame based on information about the data, the user’s interaction with the data, and/or the user’s schedule.

Identifying Listed Properties Based On Imagery, Charles Armstrong

Defensive Publications Series

Generally, the present disclosure is directed to identifying properties that are listed for sale or similarly available based on imagery of the properties. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict that a property is listed for sale or similarly available and/or characteristics about the listing based on imagery of the property, including imagery containing listing signage.

Machine-Learning Predicted User Interfaces, David Huffaker

Defensive Publications Series

Generally, the present disclosure is directed to customizing user interfaces to a particular user. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict the likelihood a user will interact with a user interface element and/or the importance of the content of the user interface element based on user data such as, for example, user interaction with the user interface.

Reinforcement Learning For Fuzzing Testing Techniques, Neil Dhillon, Tanmay Wadhwa

Defensive Publications Series

Generally, the present disclosure is directed to using machine learning to manage a trade-off between exploration and exploitation in a fuzzer system. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict inputs that will cause a system being tested to malfunction or crash based on input data provided to the system and output from the system.

Improving Appeal Of An Advertisement Based On Linguistic Trends, Donny Greenberg

Defensive Publications Series

Generally, the present disclosure is directed to improving appeal of an advertisement based on linguistic trends. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to generate fashionable advertisements based on a candidate advertisement and linguistic trends determined from sample material from one or more advertising platforms.

A Modeling System For Integral Simulation Of Propagation Of Ocean Surge And Wave And Their Impinging On Coastal Structure (Extended Abstract), Ke Qu, Hansong Tang, Anil Agrawal

Ocean Waves Workshop

Now it has become necessary to develop our capability to directly simulate many emerging coastal ocean flow and wave problems. These flow problems present a common challenge to our modeling capability; they involve multiphysics phenomena spanning a vast range of spatial and temporal scales, however, so far essentially we have no methods and computer software to directly and integrally simulate these phenomena. Now it has become necessary for us to develop new capability to model them in efficient ways and high-fidelity. Towards this goal, we have developed a brand new, unprecedented modeling system that is able to directly evaluate many …

Session 1 Paper - Contemporary Expeditionary Warfare For Scientists And Engineers, C. Reid Nichols

Ocean Waves Workshop

Expeditionary forces from many countries have deployed to mitigate the effects of natural disasters and human conflicts. Approaches are usually linked to the nation’s maritime strategy. Expeditionary forces provide rapid response to man-made disruptions such as mass casualties by terrorists, vessel hijackings by pirates, murders by drug cartels, and nuclear accidents. Recent examples include Humanitarian Assistance/Disaster Response (HA/DR) and Non Combatant Evacuation Operations (NEO). HA/DR type missions have been planned and executed to mitigate the effects of catastrophic tropical cyclones, flooding, wild fires, and tsunami. NEO missions are focused on the evacuation of a nation’s citizens and other foreigners from …

Full Of Indigo, Owen Medgyesy

The Anthology

No abstract provided.

Configuring Alarm System Based On Time To Arrive At Appointment, Nick Felker

Defensive Publications Series

Generally, the present disclosure is directed to setting an alarm to alert a user based on an appointment. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict a time to alert a user based on a user’s schedule and location.

Predictive Cryptocurrency Mining And Staking, Thomas Price

Defensive Publications Series

Generally, the present disclosure is directed to determining the validity of a chain within a blockchain system. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict the likelihood that a block within a blockchain system will be verified based on characteristics of the block.

Predicting Computing Prices Dynamically Using Machine Learning, Thomas Price

Defensive Publications Series

Generally, the present disclosure is directed to determining the cheapest datacenter for a computing task to be computed in. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict the cost of computing at a datacenter over a time interval based on time data, weather data, and/or grid statistics.

Predicting Delivery Time Of Components In A Supply Chain, Venki Raman, Sunil Reddy

Defensive Publications Series

Generally, the present disclosure is directed to predicting a delivery time of components in a supply chain. In particular, in some implementations, the systems and methods of the present disclosure can include or otherwise leverage one or more machine-learned models to predict a delivery time of a component based on enterprise data relating to the component.