Learning perceptually grounded word meanings from unaligned parallel data

In order for robots to effectively understand natural language commands, they must be able to acquire meaning representations that can be mapped to perceptual features in the external world. Previous approaches to learning these grounded meaning representations require detailed annotations at training time. In this paper, we present an approach to grounded language acquisition which is capable of jointly learning a policy for following natural language commands such as “Pick up the tire pallet,” as well as a mapping between specific phrases in the language and aspects of the external world; for example the mapping between the words “the tire pallet” and a specific object in the environment. Our approach assumes a parametric form for the policy that the robot uses to choose actions in response to a natural language command that factors based on the structure of the language. We use a gradient method to optimize model parameters. Our evaluation demonstrates the effectiveness of the model on a corpus of commands given to a robotic forklift by untrained users.     

Deficits on the continuous performance test within the schizophrenia spectrum and the mediating effects of family history of schizophrenia.

Individuals with schizophrenia spectrum personality disorders (SSPD) and schizophrenia show similar cognitive impairments. The authors examined the contributions of SSPD symptoms and familial risk for schizophrenia to impairments on the Continuous Performance Test-Identical Pairs Version. Participants included 103 schizophrenia patients, 66 first-degree relatives (29 SSPD), and 103 community controls (26 SSPD) screened for family history of psychosis. Patients and SSPD relatives performed significantly worse than non-SSPD relatives and SSPD and non-SSPD community controls. No differences in performance were observed among non-SSPD relatives and SSPD and non-SSPD community controls. Results suggest a continuum in which risk for schizophrenia-related cognitive impairments is highest among patients and SSPD relatives, intermediate among non-SSPD relatives, and lowest among SSPD and non-SSPD community controls. Results suggest that SSPD in the absence of a family history of psychosis may be a phenocopy. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Glycogen Synthase Kinase 3β Inhibition as a Therapeutic Approach in the Treatment of Endometrial Cancer

Alternative strategies beyond current chemotherapy and radiation therapy regimens are needed in the treatment of advanced stage and recurrent endometrial cancers. There is considerable promise for biologic agents targeting the extracellular signal-regulated kinase (ERK) pathway for treatment of these cancers. Many downstream substrates of the ERK signaling pathway, such as glycogen synthase kinase 3β (GSK3β), and their roles in endometrial carcinogenesis have not yet been investigated. In this study, we tested the importance of GSK3β inhibition in endometrial cancer cell lines and in vivo models. Inhibition of GSK3β by either lithium chloride (LiCl) or specific GSK3β inhibitor VIII showed cytostatic and cytotoxic effects on multiple endometrial cancer cell lines, with little effect on the immortalized normal endometrial cell line. Flow cytometry and immunofluorescence revealed a G2/M cell cycle arrest in both type I (AN3CA, KLE, and RL952) and type II (ARK1) endometrial cancer cell lines. In addition, LiCl pre-treatment sensitized AN3CA cells to the chemotherapy agent paclitaxel. Administration of LiCl to AN3CA tumor-bearing mice resulted in partial or complete regression of some tumors. Thus, GSK3β activity is associated with endometrial cancer tumorigenesis and its pharmacologic inhibition reduces cell proliferation and tumor growth.     

The Impact of Silicon on Photosynthetic and Biochemical Responses of Sugarcane under Different Soil Moisture Levels

Silicon - Increasing drought stress is one of the most limiting factors for agricultural crop productivity across the world. Silicon (Si) has been known to augment plant protection against drought...     

Epoxidized rice bran oil (ERBO) as a plasticizer for poly(vinyl chloride) (PVC)

With environmental and toxicity concerns becoming more critical, there are increasing efforts to remove phthalates from polymer compounds around the globe more rapidly. Phthalates can be replaced by natural products; in particular, those obtained from vegetable oils and fats. In the present study, a natural-based plasticizer, synthesized by epoxidation of non-toxic rice bran oil (RBO) with peroxy acid generated in situ has been added to poly(vinyl chloride). The influence of various reaction parameters on epoxidation was studied by investigating the reaction ratio, temperature, reaction time and stirring speed. Epoxidized rice bran oil (ERBO) obtained from an optimized reaction condition was analyzed by iodine number and oxirane content. FTIR was used to analyze epoxy group formation. Product ERBO was obtained with 82 % oxirane conversion within 3 h of reaction period. PVC sheets were formulated using a conventional plasticizer di-(octyl) phthalate and was partially replaced by synthesized ERBO. The effect of ERBO addition was tested by mechanical properties (tensile strength, modulus, elongation-at-break, shore D hardness) and compared with commercially available ESBO (epoxidized soybean oil). ERBO presented fairly good incorporation and plasticizing performance, as demonstrated by the results of mechanical properties, exudation, migration tests, thermal stability by thermogravimetric analysis, T _g values as shown by differential scanning calorimetry, replacing about 60 % of the total plasticizer.     

In situ tuning of graphene oxide morphology by electrochemical exfoliation

Journal of Materials Science - This paper illustrates the electrochemical exfoliation of graphite for the synthesis of graphene oxide (GO) with few layers. Innovative design and arrangement of...     

Advanced PIV/LIF and shadowgraphy system to visualize flow structure in two-phase bubbly flows

Particle image velocimetry (PIV) is a promising technique to measure dispersed phase size, dispersed phase hold-up and velocity of both the phases. The current work reports measurement of the shape, size, velocity and acceleration of bubbles using shadowgraphy, and liquid velocity measurement obtained using PIV/LIF with fluorescent tracer particles. Measurements were performed in a narrow rectangular column at moderate gas hold-up (～5%) with wide variation of bubble sizes (0.1–15 mm). The liquid velocity field was subjected to 2D discrete wavelet transform (DWT) to visualize the flow structures in the bubbly flow. Further, the slip velocity of individual bubbles was obtained from the DWT filtered liquid velocity field. The results are compared with the slip velocity correlations reported in literature for single bubbles rising in quiescent water. The comparison shows the difference in slip velocity of single bubbles and bubbles rising in swarm. The scale wise decomposition obtained from DWT was also used to quantify the liquid velocity field in terms of wavenumber spectrum. The velocity and acceleration measurements are demonstrated on a single spherical cap bubble rising in quiescent water. The measurements show the potential of the 2D acceleration measurement to facilitate the estimation of unsteady drag on bubbles.     

Biosynthesis of the Fluorinated Natural Product Nucleocidin in Streptomyces calvus Is Dependent on the bldA‐Specified Leu‐tRNAUUA Molecule

Nucleocidin is one of the very few natural products known to contain fluorine. Mysteriously, the nucleocidin producer Streptomyces calvus ATCC 13382 has not been observed to synthesize the compound since its discovery in 1956. Here, we report that complementation of S. calvus ATCC 13382 with a functional bldA‐encoded Leu‐tRNA^UUA molecule restores the production of nucleocidin. Nucleocidin was detected in culture extracts by ¹⁹F NMR spectroscopy, HPLC‐ESI‐MS, and HPLC‐continuum source molecular absorption spectroscopy for fluorine‐specific detection. The molecule was purified from a large‐scale culture and definitively characterized by NMR spectroscopy and high‐resolution MS. The nucleocidin biosynthetic gene cluster was identified by the presence of genes encoding the 5′‐O‐sulfamate moiety and confirmed by gene disruption. Two of the genes within the nucleocidin biosynthetic gene cluster contain TTA codons, thus explaining the dependence on bldA and resolving a 60‐year‐old mystery.     

Fused Deposition Modeling of Polyolefins: Challenges and Opportunities

Polyolefins are the largest class of commercially available synthetic polymers that are extensively used in a variety of applications from commodities to engineering owing to their low cost of production, good physico-mechanical properties, light weight, good processability, and recyclability. Compared to conventional molding techniques, fused deposition modeling (FDM)-based 3D printing is a smart manufacturing technology for thermoplastics due to its low cost, ease of production of complex geometrical parts, rapid prototyping, and scalable customization. FDM 3D printing can be an ideal manufacturing technology for polyolefins to manufacture various complex parts. However, FDM 3D-printing of polyolefins is challenged bycritical printing problems like high warpage, dimensional inaccuracies, poor bed adhesion, and poor layer-to-layer adhesion. In this review, a fundamental understanding of polyolefins and their FDM 3D-printing process is established, and the recent progress of FDM 3D printing of polyolefins is summarized. Furthermore, strategies to overcome warpage and to improve mechanical strength of the 3D-printed polyolefins are provided. Finally, future prospectives of FDM 3D-printing of polyolefins are critically discussed to inspire prospective research in this field. It is believed that this review article can be tremendously useful for research work related to FDM of polyolefin-based materials.