COMPRESSIVE SCANNING SPECTROSCOPY
Mask-modulated spectra are incident to a sensor and are summed during a frame time. After the frame time, a compressed spectrum is read out based on the sum and decompressed to obtain spectra for some or all specimen locations. The mask-modulated spectrum that are summed are associated with different modulations produced by a common mask.
MACHINE LEARNING MODEL EXPLANATION APPARATUS AND METHODS
Our invention is a generalizable approach for understanding machine learning models using the model inputs and outputs only. We leverage topological data analysis (TDA), which is a new field for understanding complex high dimensional data by simplifying it into human understandable shapes capturing the most salient structures in the data. Our analysis approach builds on the 'Mapper" technique in a novel way by designing a cover specifically for machine learning predictions. This cover scheme substantially differentiates our approach from the existing Mapper algorithm. This allows us to overcome the scalability limitations of Mapper, which require the cover to be low dimensional. While our approach is tailored specifically towards machine learning applications, it scales to many dimensions, i.e., prediction classes. Additionally, we developed an analysis approach called 'escape routes" to explain relationships between different regions in the topological space defined model predictions.
Search systems and computer-implemented search methods
Search systems and computer-implemented search methods are described. In one aspect, a search system includes a communications interface configured to access a plurality of data items of a collection, wherein the data items include a plurality of image objects individually comprising image data utilized to generate an image of the respective data item. The search system may include processing circuitry coupled with the communications interface and configured to process the image data of the data items of the collection to identify a plurality of image content facets which are indicative of image content contained within the images and to associate the image objects with the image content facets and a display coupled with the processing circuitry and configured to depict the image objects associated with the image content facets.
FEATURE IDENTIFICATION OR CLASSIFICATION USING TASK-SPECIFIC METADATA
Innovations in the identification or classification of features in a data set are described, such as a data set representing measurements taken by a scientific instrument. For example, a task-specific processing component, such as a video encoder, is used to generate task-specific metadata. When the data set includes video frames, metadata can include information regarding motion of image elements between frames, or other differences between frames. A feature of the data set, such as an event, can be identified or classified based on the metadata. For example, an event can be identified when metadata for one or more elements of the data set exceed one or more threshold values. When the feature is identified or classified, an output, such as a display or notification, can be generated. Although the metadata may be useable to generate a task-specific output, such as compressed video data, the identifying or classifying is not used solely in production of, or the creation of an association with, the task-specific output.
SNAP-IN BUSHINGS AND PROCESS FOR HIGH-PRESSURE AND/OR HIGH TEMPERATURE MAGIC ANGLE SPINNING NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY
Snap-in bushings are disclosed that enable sealing of sample chambers in MAS-NMR rotors for high pressure and/or high temperature operation that enhance pressure limits up to about 400 bar and temperature limits up to at least about 250.degree. C.
POROUS MATERIALS FOR ENHANCED IONIZATION EFFICIENCY (iEdison No. 0685901-19-0013)
We have employed established porous engineered materials, such as metal-organic frameworks (MOFs) to enhance the ionization efficiency and simplify sample loading--greatly expanding our capabilities in low level characterization of environmentally and/or strategically important elements.
SYSTEMS AND METHODS FOR SEPARATING YTTRIUM AND STRONTIUM (iEdison No. 0685901-17-0017)
PNNL has developed a method for the production of high-purity 90Y from 90Sr stocks. The method was designed such that it could be fully automated. The objective of the development work was to devise a process that could purify multi-Ci levels of 90Y. As this level of radiation would require the use of hot cells for production, automation is a highly desirable feature. Also, since the radiotoxicity of 90Sr is so great (bone seeker, 30 year half-life), the degree of chemical purity of the 90Y is immense ( > /= 1x10^6 degrees of purity). Therefore, PNNL designed a tandem column purification process, which allows the 90Y to be purified across two columns placed in tandem. PNNL surveyed a wide range of inorganic sorbents for use in the 90Y purification process. In previous research (Lynntech project and associated PNNL patent (attached)), we found that an antimony silicate resin possessed good properties for holding up 90Sr while allowing 90Y to pass through (though inefficiently) in a HCl matrix. However, for multi-Ci levels of 90Y production, a generator system with 90Sr affixed to the inorganic sorbent would degrade quickly due to radiolysis. Hence, PNNL developed a new type of generator, wherein the 90Sr is contained in a "liquid cow" versus a "column-bound cow". In this manner, the radioactivity would be stored outside of the purification system, thus minimizing radiolytic degradation of the system. Tandem column generator -- PRIMARY COLUMN LOAD: A recent Chinese publication (attached) described a method wherein antimony silicate nanocrystals could be formed. The researchers then characterized the 90Sr sorption properties onto these nanocrystals in HCl media. Since nanocrystals cannot be used in a generator column, PNNL extrapolated that basic premise of the reaction and used it to co-form antimony silicate nanocrystals into / onto a silica gel support (which could be used in a column format). Through parametric optimization of the synthesis process, we determined that an antimony silicate resin could be formed without incorporation of these nanocrystals on the Si gel surface. Rather, we were able to form an antimony silicate complex directly on the Si gel surface structures. This approach yielded the best separation factors for 90Y from 90Sr. Furthermore, PNNL needed to use the antimony silicate resin in "reverse" of the configuration that was used in the earlier patent (and also described in the recent Chinese publication). Instead of 90Sr being bound to the resin and 90Y passing through, we needed a process wherein the 90Sr would pass through the resin and temporarily retain the 90Y (this is necessary to avoid extreme radiolysis of multi-Ci 90Sr sources). We found that in concentrated formic acid, 90Sr has no affinity for the resin and 90Y has a good affinity for it. Therefore, a 90Sr stock solution can be passed through the resin column to extract all the 90Y, and the 90Sr stock solution can be cycled back to its storage container. Tandem column generator -- PRIMARY COLUMN TRANSFER TO SECONDARY COLUMN: PNNL determined that a dilute solution of HCl was effective in removing the sorbed 90Y from the antimony silicate column. Additionally, we determined that a tiny column of HDEHP-impregnated chromatography resin (sold as Ln Resin by Eichrom Technologies) would adsorb 90Y in dilute HCl. Therefore, the dilute HCl solution could be used to "transfer" 90Y from the primary column to the secondary column. Importantly, any trace / residual 90Sr that would pass to the secondary column would not be retained on the secondary column. Hence, the method has two degrees of purification. Tandem column generator -- SECONDARY COLUMN ELUTION: PNNL determined that stronger HCl (1 to > 6 M) was highly effective at eluting 90Y from the secondary column. Virtually all the 90Y could be eluted from the secondary column in
COMPRESSIVE TRANSMISSION MICROSCOPY
Transmission microscopy imaging systems include a mask and/or other modulator situated to encode image beams, e.g., by deflecting the image beam with respect to the mask and/or sensor. The beam is modulated/masked either before or after transmission through a sample to induce a spatially and/or temporally encoded signal by modifying any of the beam/image components including the phase/coherence, intensity, or position of the beam at the sensor. For example, a mask can be placed/translated through the beam so that several masked beams are received by a sensor during a single sensor integration time. Images associated with multiple mask displacements are then used to reconstruct a video sequence using a compressive sensing method. Another example of masked modulation involves a mechanism for phase-retrieval, whereby the beam is modulated by a set of different masks in the image plane and each masked image is recorded in the diffraction plane.
FEATURE IDENTIFICATION OR CLASSIFICATION USING TASK-SPECIFIC METADATA
Innovations in the identification or classification of features in a data set are described, such as a data set representing measurements taken by a scientific instrument. For example, a task-specific processing component, such as a video encoder, is used to generate task-specific metadata. When the data set includes video frames, metadata can include information regarding motion of image elements between frames, or other differences between frames. A feature of the data set, such as an event, can be identified or classified based on the metadata. For example, an event can be identified when metadata for one or more elements of the data set exceed one or more threshold values. When the feature is identified or classified, an output, such as a display or notification, can be generated. Although the metadata may be useable to generate a task-specific output, such as compressed video data, the identifying or classifying is not used solely in production of, or the creation of an association with, the task-specific output.
PERFORMANCE AND USABILITY ENHANCEMENTS FOR CONTINUOUS SUBGRAPH MATCHING QUERIES ON GRAPH-STRUCTURED DATA
StreamWorks is a network analysis framework that enables an analyst to monitor and analyze streaming computer network traffic data to identify emerging computer network intrusions and threats. Different types of graphical query patterns may be defined for specific types of cyberattacks including various network scans, reflector attacks, flood attack, viruses, worms, etc. StreamWorks will support subgraph matching on computer network attributes such as hostnames, IP addresses, protocols, ports, packet sizes, machine types, and message types. The speed of subgraph pattern matching will be accelerated by collecting and utilizing node and edge frequency information to optimize search paths through a massive data graph. Computer network intrusion analysis will involve live computer network data streamed in at high data rates and the analysis of data graphs consisting of millions to billions of edges. For known patterns, specific graphical query patterns are collected in a library and continuously and efficiently matched against the dynamic graph as it is updated. Each graph query is captured as a subgraph join tree which decomposes the query graph into smaller search subpatterns. These smaller subpatterns signify precursor events that emerge early before the full query pattern is complete. As precursor events are detected in data streams, they are matched to the nodes of different subgraph join trees. Matching that occurs higher in a join tree indicates a higher probability that a specific type of attack is occurring. A similarity or confidence score may be computed for the partial matching through training on collected computer network traffic data to measure the frequencies of occurrence of partial subpatterns as precursors to the full graph query pattern. For unknown patterns or zero-day exploits, the same analysis framework may be applied to track the emergence of small subpatterns as they appear in the data stream. The system may be seeded with hints to look for small graph patterns that involve rare events (based on collected statistics), events involving critical resources such as an authentication server, domain name server, database, etc., or particular host machines of specific suspicions or interests to analysts. When seeded subpatterns are found in the data stream, they are tracked and monitored within subgraph join trees. Here, subpatterns are joined based on specific criteria such as when the subpatterns grow beyond a certain threshold size, additional critical resources are introduced into a subpattern, or important types of interactions or communications are detected. Thus, full attack patterns may dynamically emerge from the small seeded patterns or hints. The initial seeded patterns may have confidence scores generated from collected statistics or assigned by analysts, which are then propagated up through the subgraph join tree. Additionally, StreamWorks will provide mechanisms for analysts to vet tracked subpatterns so as to improve analysis and performance by eliminating benign patterns from being monitored and assessed. The advanced dynamic graph algorithms have been packaged into a streaming network analysis framework known as StreamWorks. With StreamWorks, a scientist or analyst may detect and identify precursor events and patterns as they emerge in complex networks. This analysis framework is intended to be used in a dynamic environment where network data is streamed in and is appended to a large-scale dynamic graph. An interactive graph query construction tool has been developed that will allow an analyst to build a query graph. Various cyberattack templates have been developed for querying the dynamic graph, where an analyst may tailor the attributes of a cyberattack query by adjusting parameters of the cyberattack template. The dynamic results, which are the subpatterns of the template that are matched in the dynamic graph, are returned to the analyst in a visualization showing the emerging and evolving patterns along with a visualization of the subgraph join tree containing statistics on the level of matching per partial subgraph pattern.