Abstract: The present disclosure provides directed to new, more efficient neural network architectures. As one example, in some implementations, the neural network architectures of the present disclosure can include a linear bottleneck layer positioned structurally prior to and/or after one or more convolutional layers, such as, for example, one or more depthwise separable convolutional layers. As another example, in some implementations, the neural network architectures of the present disclosure can include one or more inverted residual blocks where the input and output of the inverted residual block are thin bottleneck layers, while an intermediate layer is an expanded representation. For example, the expanded representation can include one or more convolutional layers, such as, for example, one or more depthwise separable convolutional layers. A residual shortcut connection can exist between the thin bottleneck layers that play a role of an input and output of the inverted residual block.

Abstract: In an example of the disclosure, an intermediate transfer member cleaning system includes an intermediate transfer member (“blanket”), an endless cleaning surface, and a pyrolysis station. The blanket is to receive a thermoplastic print agent from a photoconductive element. The endless cleaning surface can be positioned to rotatably engage with the blanket to transfer a residue of the thermoplastic print agent from the blanket to the endless cleaning surface. The endless cleaning surface can be moved away from the blanket to enter a pyrolysis station. The pyrolysis station is for receiving the endless cleaning surface, for heating the endless cleaning surface to convert the residue to ash, and for removing the ash.

Abstract: In one example, an exhaust hood includes an enclosure having a perimeter defining an exhaust area and a fluid flow path. The fluid flow path includes a perimeter intake slot through which air is sucked into the flow path during an exhaust operation and a flow channel in fluid communication with the perimeter intake slot and configured to carry fluid away from the intake slot and out of the enclosure.

Abstract: The present disclosure provides directed to new, more efficient neural network architectures. As one example, in some implementations, the neural network architectures of the present disclosure can include a linear bottleneck layer positioned structurally prior to and/or after one or more convolutional layers, such as, for example, one or more depthwise separable convolutional layers. As another example, in some implementations, the neural network architectures of the present disclosure can include one or more inverted residual blocks where the input and output of the inverted residual block are thin bottleneck layers, while an intermediate layer is an expanded representation. For example, the expanded representation can include one or more convolutional layers, such as, for example, one or more depthwise separable convolutional layers. A residual shortcut connection can exist between the thin bottleneck layers that play a role of an input and output of the inverted residual block.

Abstract: A printing apparatus has a transfer member to receive an image and transfer the image to a substrate, and an airflow management unit disposed adjacent to the transfer member to define a channel between the transfer member and the airflow management unit. The airflow management unit has an injection mechanism to inject airflow into the channel, a suction mechanism to remove the airflow from the channel, and a vortex generator to generate a vortex within the channel.

Abstract: Described herein is a liquid electrophotographic photovoltaic ink composition comprising: a dispersion of a material with a perovskite structure, a thermoplastic resin and conductive particles in a carrier liquid; wherein the material with a perovskite structure has a chemical formula selected from ABX3 and A2BX6; wherein A is a cation, B is a cation and X is an anion; and wherein the thermoplastic resin comprises: a copolymer of an alkylene monomer and a monomer having acidic side groups; and/or a copolymer of an alkylene monomer and an ethylenically unsaturated monomer comprising an epoxide; and/or a copolymer of an alkylene monomer, an ethylenically unsaturated monomer comprising an epoxide, and a monomer selected from a monomer having acidic side groups, a monomer having ester side groups and a mixture thereof. Also described is a method of producing a photovoltaic cell using the LEP ink and the printed cell produced by the method.

Abstract: The present disclosure provides a new type of generalized artificial neural network where neurons and synapses maintain multiple states. While classical gradient-based backpropagation in artificial neural networks can be seen as a special case of a two-state network where one state is used for activations and another for gradients with update rules derived from the chain rule, example implementations of the generalized framework proposed herein may additionally: have neither explicit notion of nor ever receive gradients; contain more than two states; and/or implement or apply learned (e.g., meta-learned) update rules that control updates to the state(s) of the neuron during forward and/or backward propagation of information.

Abstract: The present disclosure provides systems and methods that enable parameter-efficient transfer learning, multi-task learning, and/or other forms of model re-purposing such as model personalization or domain adaptation. In particular, as one example, a computing system can obtain a machine-learned model that has been previously trained on a first training dataset to perform a first task. The machine-learned model can include a first set of learnable parameters. The computing system can modify the machine-learned model to include a model patch, where the model patch includes a second set of learnable parameters. The computing system can train the machine-learned model on a second training dataset to perform a second task that is different from the first task, which may include learning new values for the second set of learnable parameters included in the model patch while keeping at least some (e.g., all) of the first set of parameters fixed.

Abstract: The present disclosure provides directed to new, more efficient neural network architectures. As one example, in some implementations, the neural network architectures of the present disclosure can include a linear bottleneck layer positioned structurally prior to and/or after one or more convolutional layers, such as, for example, one or more depthwise separable convolutional layers. As another example, in some implementations, the neural network architectures of the present disclosure can include one or more inverted residual blocks where the input and output of the inverted residual block are thin bottleneck layers, while an intermediate layer is an expanded representation. For example, the expanded representation can include one or more convolutional layers, such as, for example, one or more depthwise separable convolutional layers. A residual shortcut connection can exist between the thin bottleneck layers that play a role of an input and output of the inverted residual block.

Abstract: According to some examples in the disclosure, a method may comprise providing a first stream of discrete volumes of material; and directing a pulsed laser beam at a first discrete volume of material in the first stream of discrete volumes of material so as to interact with the first discrete volume of material and thereby displace the first discrete volume away from the first stream. An apparatus and a system are also disclosed.

Abstract: An example method comprises identifying, by a processor, a gap in a housing of a printing apparatus and positioning an inlet of a conduit inside the housing proximate the gap. The inlet is fluidly connected to a fan, and the fan is powered to create a pressure differential across the housing to minimize the amount of air inside the housing being able to escape the housing via the gap.

Abstract: The present disclosure provides systems and methods that enable parameter-efficient transfer learning, multi-task learning, and/or other forms of model re-purposing such as model personalization or domain adaptation. In particular, as one example, a computing system can obtain a machine-learned model that has been previously trained on a first training dataset to perform a first task. The machine-learned model can include a first set of learnable parameters. The computing system can modify the machine-learned model to include a model patch, where the model patch includes a second set of learnable parameters. The computing system can train the machine-learned model on a second training dataset to perform a second task that is different from the first task, which may include learning new values for the second set of learnable parameters included in the model patch while keeping at least some (e.g., all) of the first set of parameters fixed.

Abstract: Disclosed is a method to mitigate deterioration of printing fluid in a tank of a printer. The method comprises: draining printing fluid from a tank of a printer, and stopping the draining before the tank is empty, wherein the printing fluid comprises solids suspended in imaging oil, filtering the printing fluid drained from the tank, to separate at least some of the solids from the imaging oil, passing the filtered imaging oil to a reservoir; and then passing imaging oil from the reservoir to the tank. Also disclosed is a printer and a system to perform the method.

Abstract: Described herein is a liquid electrophotographic photovoltaic ink composition comprising: a dispersion of a material with a perovskite structure, a thermoplastic resin and conductive particles in a carrier liquid; wherein the material with a perovskite structure has a chemical formula selected from ABX3 and A2BX6; wherein A is a cation, B is a cation and X is an anion; and wherein the thermoplastic resin comprises: a copolymer of an alkylene monomer and a monomer having acidic side groups; and/or a copolymer of an alkylene monomer and an ethylenically unsaturated monomer comprising an epoxide; and/or a copolymer of an alkylene monomer, an ethylenically unsaturated monomer comprising an epoxide, and a monomer selected from a monomer having acidic side groups, a monomer having ester side groups and a mixture thereof. Also described is a method of producing a photovoltaic cell using the LEP ink and the printed cell produced by the method.

Abstract: An example system includes a conveyor. The conveyor is not a photoconductor. The system also includes a developer unit. The developer unit is to internally concentrate printing liquid. The developer unit also is to deliver the printing liquid to the conveyor. The system includes a wiper in contact with the conveyor. The wiper is to remove the printing liquid from the conveyor.

Abstract: A print apparatus is disclosed. The print apparatus comprises a container-receiving unit for receiving a reusable print agent container; a print agent reservoir for storing print agent to be consumed by the print apparatus during a printing operation; a pump for transferring print agent between a reusable print agent container positioned in the container-receiving unit and the print agent reservoir; and processing circuitry. The processing circuitry is to operate the pump to transfer print agent of a first concentration from the reusable print agent container to the print agent reservoir; and, responsive to determining that a volume of print agent of the first concentration in the reusable print agent container has fallen below a threshold volume, operate the pump to transfer a volume of print agent of a second, lower concentration from the print agent reservoir to the reusable print agent container. A method and a machine-readable medium are also disclosed.

Abstract: In one example of the disclosure, a first transfer of ink is made from a photoconductor to a blanket in contact with the photoconductor. The blanket is to cycle along a path. The first transfer occurs at a first arc of the blanket path. A second transfer of the ink is made from the blanket to a media in contact with the blanket. The second transfer occurs at a second arc of the blanket path. A heat source located adjacent to a third arc of the blanket path is utilized to heat an external surface of the blanket. The heating is to occur following the second transfer of the ink.

Abstract: Disclosed is a method to mitigate deterioration of printing fluid in a tank of a printer. The method comprises: draining printing fluid from a tank of a printer, and stopping the draining before the tank is empty, wherein the printing fluid comprises solids suspended in imaging oil, filtering the printing fluid drained from the tank, to separate at least some of the solids from the imaging oil, passing the filtered imaging oil to a reservoir; and then passing imaging oil from the reservoir to the tank. Also disclosed is a printer and a system to perform the method.

Abstract: Herein is disclosed a method for concentrating a liquid electrostatic ink composition, the method comprising replacing a first liquid carrier in a liquid electrostatic ink composition with a second liquid carrier, wherein the liquid electrostatic ink composition comprises chargeable toner particles dispersed in the first liquid carrier; and removing the second liquid carrier to produce a concentrated liquid electrostatic ink composition, wherein the second liquid carrier has a boiling point below 100° C. and wherein the second liquid carrier has a lower boiling point than the first liquid carrier. Also disclosed herein is a concentrated liquid electrostatic ink composition producible according to the method.

Abstract: An example method comprises identifying, by a processor, a gap in a housing of a printing apparatus and positioning an inlet of a conduit inside the housing proximate the gap. The inlet is fluidly connected to a fan, and the fan is powered to create a pressure differential across the housing to minimize the amount of air inside the housing being able to escape the housing via the gap.