Abstract: An electronic device may include a lens module with a tunable lens. User comfort when operating the electronic device may be improved by mitigating changes in magnification at different optical powers of the lens module. One technique to mitigate changes in magnification at different optical powers is to physically move a tunable lens in the lens module relative to a non-tunable lens in the lens module to change a gap between the tunable lens and the non-tunable lens. Another technique is to include two tunable lenses such as an adjustable positive optical power lens and an adjustable negative optical power lens in the lens module. Instead or in addition, adjustments to a tunable lens in a lens module may be performed during one or more eye blinks and/or over a transition period.

Abstract: Methods and apparatuses, including systems, for forming mycelium fabrics, manufacturing and/or production in large-scale mushroom farm facilities (e.g., commercial scale under an agroecological approach). These methods and apparatuses may allow the scaling-up of manufacturing of mycotextiles in a manner that is both cost-effective and respectful of the environment, including minimizing the production of harmful byproducts. These methods and apparatuses include adding a stabilized formulation containing growth inducers and functionalized nanoparticles to allow both fungal mycelium growth and in situ and in vivo nanocrosslinking.

Abstract: Described herein are oil-in-water (o/w) nanoemulsions for internal humectation of mycelium-based textiles using fatty acids of vegetable origin and/or their synthetic derivatives and non-ionic surfactants. These nanoemulsions may be used as fatliquoring solutions in materials based on fungal mycelium, bacterial recombinant proteins, plant-based textiles, and/or other materials used as animal leather substitutes, including its potential use as environmentally friendly (e.g., “green”) and cost-effective fatliquoring alternative for animal leather. Also described are methods and apparatuses for scaling up, recovering, and reusing nanoemulsions. The nanoemulsion fatliquoring compositions and methods described herein may reduce frictional forces, improve tensile strength and elongation at break, introduce softness, and protect mycelium-based textile material against cracking.

Abstract: Systems and methods to mimic a target food item using artificial intelligence are disclosed. The system can learn from open source and proprietary databases. A prediction model can be trained using features of the source ingredients to match those of the given target food item. A formula comprising a combination of most relevant source ingredients and their proportions can be determined using the trained prediction model. A set of existing recipes can be used as a dataset to train a recurrent neural network (RNN) and/or other suitable models. The RNN can be used to determine a recipe to mimic the target food item. The recipe may comprise a cooking process for the set of ingredients in the formula and can be cooked by a chef. The recipe may be further modified as necessary based on human feedback on sensorial descriptors.

Abstract: An electronic device may include an optical module with a display and a tunable lens. During operation, the electronic device may gather data and adjust the tunable lens based on the gathered data. The optical module may include a non-adjustable lens element with convex curvature in addition to the tunable lens. The optical module may include a Fresnel lens element in addition to the tunable lens. The optical module may include a catadioptric lens in addition to the tunable lens. The optical module may include a catadioptric lens that includes the tunable lens. The optical module may have a birdbath architecture that includes the tunable lens. The optical module may include a waveguide and the tunable lens may be an adjustable positive bias lens and/or an adjustable negative bias lens.

Abstract: An electronic device may include a display panel configured to produce light and a lens assembly that receives the light from the display panel. The lens assembly may be a catadioptric lens assembly with a partially reflective layer, lens element, quarter wave plate, and reflective polarizer. Some of the light emitted by the display panel may be reflected by the catadioptric lens assembly away from the catadioptric lens assembly. A supplemental mirror may be used to reflect this light back towards the catadioptric lens assembly. The supplemental mirror may increase the field-of-view of images viewable through the catadioptric lens assembly. The supplemental mirror may have a dithered upper surface or a phase-shift mask.

Abstract: Mycotextiles, methods of making them, methods of processing them, and compositions and apparatuses for making and/or processing them are described herein.

Abstract: Mycotextiles, methods of making them, methods of processing them, and compositions and apparatuses for making and/or processing them are described herein.

Abstract: Mycotextiles, methods of making them, methods of processing them, and compositions and apparatuses for making and/or processing them are described herein.

Abstract: Techniques to generate a flavor profile using artificial intelligence are disclosed. A flavor classifier classifies flavors for a given set of ingredients of a recipe from a set of possible classes of flavors. The flavor classifier may use different deep learning models to allow for different granularity levels corresponding to each flavor based on desired preciseness with classification of a particular flavor. A respective flavor predictor may or may not be used for each granularity level based on output of a certainty level classifier used for determining a preceding level of granularity.

Abstract: Systems and methods to mimic a target food item using artificial intelligence are disclosed. The system can learn from open source and proprietary databases. A prediction model can be trained using features of the source ingredients to match those of the given target food item. A formula comprising a combination of most relevant source ingredients and their proportions can be determined using the trained prediction model. A set of existing recipes can be used as a dataset to train a recurrent neural network (RNN) and/or other suitable models. The RNN can be used to determine a recipe to mimic the target food item. The recipe may comprise a cooking process for the set of ingredients in the formula and can be cooked by a chef. The recipe may be further modified as necessary based on human feedback on sensorial descriptors.

Abstract: Systems and methods to mimic a target food item using artificial intelligence are disclosed. The system can learn from open source and proprietary databases. A prediction model can be trained using features of the source ingredients to match those of the given target food item. A formula comprising a combination of most relevant source ingredients and their proportions can be determined using the trained prediction model. A set of existing recipes can be used as a dataset to train a recurrent neural network (RNN) and/or other suitable models. The RNN can be used to determine a recipe to mimic the target food item. The recipe may comprise a cooking process for the set of ingredients in the formula and can be cooked by a chef. The recipe may be further modified as necessary based on human feedback on sensorial descriptors.

Abstract: Techniques to generate a flavor profile using artificial intelligence are disclosed. A flavor classifier classifies flavors for a given set of ingredients of a recipe from a set of possible classes of flavors. The flavor classifier may use different deep learning models to allow for different granularity levels corresponding to each flavor based on desired preciseness with classification of a particular flavor. A respective flavor predictor may or may not be used for each granularity level based on output of a certainty level classifier used for determining a preceding level of granularity.

Abstract: A color predictor is provided to predict the color of a food item given its formula comprising the ingredients and its quantities. The color predictor may utilize machine learning algorithms and a set of recipe data to train the color predictor. The color predictor can also be used by a color recommender to recommend changes in the given formula to achieve a target color.

Abstract: Techniques to generate a flavor profile using artificial intelligence are disclosed. A flavor classifier classifies flavors for a given set of ingredients of a recipe from a set of possible classes of flavors. The flavor classifier may use different deep learning models to allow for different granularity levels corresponding to each flavor based on desired preciseness with classification of a particular flavor. A respective flavor predictor may or may not be used for each granularity level based on output of a certainty level classifier used for determining a preceding level of granularity.

Abstract: A color predictor is provided to predict the color of a food item given its formula comprising the ingredients and its quantities. The color predictor may utilize machine learning algorithms and a set of recipe data to train the color predictor. The color predictor can also be used by a color recommender to recommend changes in the given formula to achieve a target color.

Abstract: A color predictor is provided to predict the color of a food item given its formula comprising the ingredients and its quantities. The color predictor may utilize machine learning algorithms and a set of recipe data to train the color predictor. The color predictor can also be used by a color recommender to recommend changes in the given formula to achieve a target color.

Abstract: A display device with one or more displays and an optical system with a plurality of channels arranged to generate an immersive virtual image from the a image, the virtual image comprising a plurality of image pixels, by each channel projecting light from the object pixels to a respective pupil range. The object pixels are grouped into clusters, each associated with a channel that produces from the object pixels a partial virtual image comprising image pixels. The clusters of at least two channels are substantially contained in opposite half-spaces defined by a plane passing by the imaginary sphere center. Each of the two channels comprises one surface on which the imaging light rays forming the partial virtual image suffer a last reflection before reaching the pupil range, where each surface is substantially contained in the opposite half-space containing their respective clusters.

Abstract: Techniques to generate a flavor profile using artificial intelligence are disclosed. A flavor classifier classifies flavors for a given set of ingredients of a recipe from a set of possible classes of flavors. The flavor classifier may use different deep learning models to allow for different granularity levels corresponding to each flavor based on desired preciseness with classification of a particular flavor. A respective flavor predictor may or may not be used for each granularity level based on output of a certainty level classifier used for determining a preceding level of granularity.

Abstract: Systems and methods to mimic a target food item using artificial intelligence are disclosed. The system can learn from open source and proprietary databases. A prediction model can be trained using features of the source ingredients to match those of the given target food item. A formula comprising a combination of most relevant source ingredients and their proportions can be determined using the trained prediction model. A set of existing recipes can be used as a dataset to train a recurrent neural network (RNN) and/or other suitable models. The RNN can be used to determine a recipe to mimic the target food item. The recipe may comprise a cooking process for the set of ingredients in the formula and can be cooked by a chef. The recipe may be further modified as necessary based on human feedback on sensorial descriptors.