Abstract: A fin field effect transistor (Fin FET) device includes a fin structure extending in a first direction and protruding from an isolation insulating layer disposed over a substrate. The fin structure includes a well layer, an oxide layer disposed over the well layer and a channel layer disposed over the oxide layer. The Fin FET device includes a gate structure covering a portion of the fin structure and extending in a second direction perpendicular to the first direction. The Fin FET device includes a source and a drain. Each of the source and drain includes a stressor layer disposed in recessed portions formed in the fin structure. The stressor layer extends above the recessed portions and applies a stress to a channel layer of the fin structure under the gate structure. The Fin FET device includes a dielectric layer formed in contact with the oxide layer and the stressor layer in the recessed portions.

Abstract: A method for manufacturing a semiconductor device includes etching a substrate to form a semiconductor fin. An isolation structure is formed above the substrate and laterally surrounds the semiconductor fin. A fin sidewall structure is formed above the isolation structure and on a sidewall of the semiconductor fin. The semiconductor fin is recessed to expose an inner sidewall of the fin sidewall structure. A source/drain epitaxial structure is grown on the recessed semiconductor fin.

Abstract: A semiconductor structure is provided. The semiconductor structure includes a gate structure, a first source/drain structure, and a contact structure. The gate structure has a gate dielectric layer over a first fin structure. The first source/drain structure is positioned in the first fin structure and adjacent to the gate structure. The first source/drain structure includes a first epitaxial layer in contact with the top surface of the first fin structure and a second epitaxial layer over the first epitaxial layer and extending above a bottom surface of the gate dielectric layer. The contact structure extends into the first source/drain structure. The top surface of the first fin structure is between a top surface and a bottom surface of the first source/drain structure.

Abstract: Certain aspects of the present disclosure provide techniques for efficiently configuring a machine learning model. An example method generally includes generating a randomly sampled data set from a data set including a larger first set of samples associated with a first classification and a smaller second set of samples associated with a second classification. An analysis plot for the machine learning model is generated based on the randomly sampled data set. A point associated with an accuracy metric for the machine learning model is identified on the analysis plot based on a slope of a line tangential to the identified point and a value identifying a relative importance of precision to recall in the machine learning model. The machine learning model is configured with a threshold value between the first classification and the second classification based at least in part on the identified point on the analysis plot.

Abstract: A fin field effect transistor (Fin FET) device includes a fin structure extending in a first direction and protruding from an isolation insulating layer disposed over a substrate. The fin structure includes a well layer, an oxide layer disposed over the well layer and a channel layer disposed over the oxide layer. The Fin FET device includes a gate structure covering a portion of the fin structure and extending in a second direction perpendicular to the first direction. The Fin FET device includes a source and a drain. Each of the source and drain includes a stressor layer disposed in recessed portions formed in the fin structure. The stressor layer extends above the recessed portions and applies a stress to a channel layer of the fin structure under the gate structure. The Fin FET device includes a dielectric layer formed in contact with the oxide layer and the stressor layer in the recessed portions.

Abstract: An integrated circuit structure includes a gate stack over a semiconductor substrate, and a silicon germanium region extending into the semiconductor substrate and adjacent to the gate stack. The silicon germanium region has a top surface, with a center portion of the top surface recessed from edge portions of the top surface to form a recess. The edge portions are on opposite sides of the center portion.

Abstract: An integrated circuit structure includes a gate stack over a semiconductor substrate, and an opening extending into the semiconductor substrate, wherein the opening is adjacent to the gate stack. A first silicon germanium region is disposed in the opening, wherein the first silicon germanium region has a first germanium percentage. A second silicon germanium region is over the first silicon germanium region. The second silicon germanium region comprises a portion in the opening. The second silicon germanium region has a second germanium percentage greater than the first germanium percentage. A silicon cap substantially free from germanium is over the second silicon germanium region.

Abstract: A semiconductor device includes a gate structure formed over a channel region of the semiconductor device, a source/drain region adjacent the channel region, and an electrically conductive contact layer over the source/drain region. The source/drain region includes a first epitaxial layer having a first material composition and a second epitaxial layer formed over the first epitaxial layer. The second epitaxial layer has a second material composition different from the first composition. The electrically conductive contact layer is in contact with the first and second epitaxial layers. A bottom of the electrically conductive contact layer is located below an uppermost portion of the first epitaxial layer.

Abstract: A device includes first and second semiconductor fins, first, second, third and fourth fin sidewall spacers, and first and second epitaxy structures. The first and second fin sidewall spacers are respectively on opposite sides of the first semiconductor fin. The third and fourth fin sidewall spacers are respectively on opposite sides of the second semiconductor fin. The first and third fin sidewall spacers are between the first and second semiconductor fins and have smaller heights than the second and fourth fin sidewall spacers. The first and second epitaxy structures are respectively on the first and second semiconductor fins and merged together.

Abstract: A static random access memory (SRAM) cell includes a semiconductor fin, a first gate structure, a second gate structure, an epitaxy structure, and a first fin sidewall structure. The first gate structure crosses the semiconductor fin to form a pull-down (PD) transistor. The second gate structure crosses the semiconductor fin to form a pull-gate (PG) transistor. The epitaxy structure is on the semiconductor fin and between the first and second gate structures. The first fin sidewall structure is on a first side of the epitaxy structure and between the first and second gate structures. A method for manufacturing the semiconductor device is also disclosed.

Abstract: A semiconductor device includes a gate structure formed over a channel region of the semiconductor device, a source/drain region adjacent the channel region, and an electrically conductive contact layer over the source/drain region. The source/drain region includes a first epitaxial layer having a first material composition and a second epitaxial layer formed over the first epitaxial layer. The second epitaxial layer has a second material composition different from the first composition. The electrically conductive contact layer is in contact with the first and second epitaxial layers. A bottom of the electrically conductive contact layer is located below an uppermost portion of the first epitaxial layer.

Abstract: A method for manufacturing an integrated circuit is provided. The method includes forming first and second semiconductor fins; forming first and second dielectric fin sidewall structures on opposite sidewalls of the first semiconductor fin, wherein the first dielectric fin sidewall structure is higher than the second dielectric fin sidewall structure, and the second dielectric fin sidewall structure is between the first and second semiconductor fins; recessing at least a portion of the first semiconductor fin between the first and second dielectric fin sidewall structures until a top of the recessed portion of the first semiconductor fin is lower than a top of the first dielectric fin sidewall structure; and forming a first epitaxy structure on the recessed portion of the first semiconductor fin.

Abstract: A fin structure on a substrate is disclosed. The fin structure can comprises a first epitaxial region and a second epitaxial region separated by a dielectric region, a merged epitaxial region on the first epitaxial region and the second epitaxial region, an epitaxial buffer region on a top surface of the merged epitaxial region, and an epitaxial capping region on the buffer epitaxial region and side surfaces of the merged epitaxial region.

Abstract: A method of forming a semiconductor device having first and second fin structures on a substrate includes forming a first epitaxial region of the first fin structure and forming a second epitaxial region of the second fin structure. The method further includes forming a buffer region on the first epitaxial region of the first fin structure and performing an etch process to etch back a portion of the second epitaxial region. The buffer region helps to prevents etch back of a top surface of the first epitaxial region during the etch process. Further, a capping region is formed on the buffer region and the etched second epitaxial region.

Abstract: A method includes forming a semiconductor fin over a substrate, etching the semiconductor fin to form a recess, wherein the recess extends into the substrate, and forming a source/drain region in the recess, wherein forming the source/drain region includes epitaxially growing a first semiconductor material on sidewalls of the recess, wherein the first semiconductor material includes silicon germanium, wherein the first semiconductor material has a first germanium concentration from 10 to 40 atomic percent, epitaxially growing a second semiconductor material over the first semiconductor material, the second semiconductor material including silicon germanium, wherein the second semiconductor material has a second germanium concentration that is greater than the first germanium concentration, and epitaxially growing a third semiconductor material over the second semiconductor material, the third semiconductor material including silicon germanium, wherein the third semiconductor material has a third germanium con

Abstract: Techniques for determining and utilizing optimal aggregation schedules are described are described. A deep machine learning model can be trained using multiple processing elements implemented in one or multiple computing devices and that are interconnected using one or multiple types of links. An optimal aggregation schedule for such arbitrary topologies can be determined automatically. The determination may include solving a linear program on the spanning tree polytope. The optimal aggregation schedule can be utilized by the multiple processing elements to train the deep machine learning model.

Abstract: A device includes a fin over a substrate, the fin including a first end and a second end, wherein the first end of the fin has a convex profile, an isolation region adjacent the fin, a gate structure along sidewalls of the fin and over the top surface of the fin, a gate spacer laterally adjacent the gate structure, and an epitaxial region adjacent the first end of the fin.

Abstract: Techniques for code-free automated machine learning (ML) are described. Users can train high-quality ML models and pipelines without necessarily needing to write code by providing a training dataset to a code-free machine learning service. The service may deploy an ML orchestration function and a storage location on behalf of a user. When a modification is made to the storage bucket, such as by the user providing a training dataset, the orchestration function is invoked and can automatically initiate an AutoML process using at least the training data to train multiple ML model variants. The resultant ML model(s) and associated metrics can be provided to the user, deployed behind an endpoint, and/or used to generate inferences.

Abstract: A method includes forming a semiconductor fin over a substrate, etching the semiconductor fin to form a recess, wherein the recess extends into the substrate, and forming a source/drain region in the recess, wherein forming the source/drain region includes epitaxially growing a first semiconductor material on sidewalls of the recess, wherein the first semiconductor material includes silicon germanium, wherein the first semiconductor material has a first germanium concentration from 10 to 40 atomic percent, epitaxially growing a second semiconductor material over the first semiconductor material, the second semiconductor material including silicon germanium, wherein the second semiconductor material has a second germanium concentration that is greater than the first germanium concentration, and epitaxially growing a third semiconductor material over the second semiconductor material, the third semiconductor material including silicon germanium, wherein the third semiconductor material has a third germanium con

Abstract: A device includes a fin over a substrate, the fin including a first end and a second end, wherein the first end of the fin has a convex profile, an isolation region adjacent the fin, a gate structure along sidewalls of the fin and over the top surface of the fin, a gate spacer laterally adjacent the gate structure, and an epitaxial region adjacent the first end of the fin.