Search for: All records

Logistic Regression is a widely used generalized linear model applied in classification settings to assign probabilities to class labels. It is also well known that logistic regression is a maximum entropy procedure subject to what are sometimes called the balance conditions. The dominant view in existing explanations are all discriminative, i.e., modeling labels given the data. This paper adds to the maximum entropy interpretation, establishing a generative, maximum entropy explanation for the commonly used logistic regression training and optimization procedures. We show that logistic regression models the conditional distribution on the instance space given class labels with a maximum entropy model subject to a first moment constraint on the training data, and that the commonly used fitting procedure would be a Monte-Carlo fit for the generative view. 

Instances of casualties resulting from large crowds persist, highlighting the existing limitations of current crowd management practices in Smart Cities. One notable drawback is the insufficient provision for disadvantaged individuals who may require additional time to evacuate due to their slower running speed. Moreover, the existing escape strategies may fall short of ensuring the safety of all individuals during a crowd surge. To address these pressing concerns, this paper proposes two crowd management methodologies. Firstly, we advocate for implementing a fair evacuation strategy following a surge event, which considers the diverse needs of all individuals, ensuring inclusivity and mitigating potential risks. Secondly, we propose a preventative approach involving the adjustment of attraction locations and switching between stage performances in large-crowded events to minimize the occurrence of surges and enhance crowd dispersion. We used high-fidelity crowd management simulators to assess the effectiveness of our proposals. Our findings demonstrate the positive impact of the fair evacuation strategy on safety measures and inclusivity, which increases fairness by 41.8% on average. Furthermore, adjusting attraction locations and stage performances has shown a significant reduction in surges by 34% on average, enhancing overall crowd safety. 

The exceptional elastic resilience of some protein materials underlies essential biomechanical functions with broad interest in biomedical fields. However, molecular design of elastic resilience is restricted to amino acid sequences of a handful of naturally occurring resilient proteins such as resilin and elastin. Here, we exploit non-resilin/elastin sequences that adopt kinetically stabilized, random coil–dominated conformations to achieve near-perfect resilience comparable with that of resilin and elastin. We also show a direct correlation between resilience and Raman-characterized protein conformations. Furthermore, we demonstrate that metastable conformation of proteins enables the construction of mechanically graded protein materials that exhibit spatially controlled conformations and resilience. These results offer insights into molecular mechanisms of protein elastomers and outline a general conformation-driven strategy for developing resilient and functional protein materials. 

Biological wastewater treatment is the process in which contaminants can be removed or degraded by various microorganisms to eliminate the negative impact on environment and human health. Given the fact that traditional physical and chemical purification methods are high-cost, unsustainable and unspecific, biotreatment is playing an increasingly important role in the wastewater treatment field. The effective implementation of biotreatment strategy relies strongly on the intrinsic degradation capability of the microorganisms as well as their interaction with pollutants. In this review, we will focus on recent technological advances in engineering and improving biotreatment at both biocatalyst and bioreactor levels. Specifically, we will discuss the progress in synthetic biology for enhancing biosorption and biotransformation, and the challenges in applying engineered microorganisms on contaminated sites. We will further review the latest developments in bioreactor design, particularly the prospects of additive manufacturing/bioprinting to further optimize the mass transport inside bioreactors through complex 3-D structures and flexible material selections. These research efforts are redefining the frontier of biotreatment, and opening up new opportunities for cost-effective, efficient, and sustainable wastewater treatment. 

Living electronics that converges the unique functioning modality of biological and electrical circuits has the potential to transform both fundamental biophysical/biochemical inquiries and translational biomedical/engineering applications. This article will review recent progress in overcoming the intrinsic physiochemical and signaling mismatches at biological/electronic interfaces, with specific focus on strategic approaches in forging the functional synergy through: (1) biohybrid electronics, where genetically encoded bio-machineries are hybridized with electronic transducers to facilitate the translation/interpretation of biologically derived signals; and (2) biosynthetic electronics, where biogenic electron pathways are designed and programmed to bridge the gap between internal biological and external electrical circuits. These efforts are reconstructing the way that artificial electronics communicate with living systems, and opening up new possibilities for many cross-disciplinary applications in biosynthesis, sensing, energy transduction, and hybrid information processing.