In todays world, many remotely operated systems are becoming ever more complex and the number of components in these systems keeps on growing. Examples include water pump systems for irrigation or mining purposes, off-shore power generation systems --the focus application area in this talk. Such systems are typically mission-critical for their users. Hence, it is of utmost importance to keep them in operating condition via well-designed maintenance policies. It makes sense for such policies to make use of whatever information is available no matter how limited. This approach is commonly labeled as condition-based maintenance (CBM) planning. In this talk, we will present a novel methodology to manage maintenance interventions as well as spare part quantity decisions for such systems. Our methodology is based on reducing the state space of the multi-component system and optimizing the resulting reduced-state Markov decision process via a linear programming approach. This methodology is highly scalable and capable of solving large problems that cannot be approached with the previously existing solution procedures due to the curse of dimensionality.

Rogers-Ramanujan identities have a central role in integer partitions. Then, they were extended combinatorially by Gordon in 1961. Andrews, found an analytic representation for Rogers-Ramanujan-Gordon identities in 1974. In this talk, a new evidently positive generating function for the difference condition side of Rogers-Ramanujan Gordon partitions when k3 is given. Also, it is combinatorially interpreted using the base partition-moves framework.

Siber Güvenlik Tezsiz Yüksek Lisans Programı Tanıtım Toplantısı 16 Nisan 2024 Salı

Veri Analitiği Yüksek Lisans Programı Tanıtım Toplantısı 16 Nisan Salı 19:00

CRISPRCas interference is mediated by Cas effector nucleases that are either components of multisubunit complexesin class1 CRISPRCas systemsor domains of a single proteinin class2 systems1,2,3. Here we show that the subtypeIII-E effector Cas7-11 is a single-protein effector in the class1 CRISPRCas systems originating from the fusion of a putative Cas11 domain and multiple Cas7 subunits that are derived from subtypeIII-D. Cas7-11 from Desulfonema ishimotonii (DiCas7-11), when expressed in Escherichia coli, has substantial RNA interference effectivity against mRNAs and bacteriophages. Similar to many class2 effectorsand unique among class1 systemsDiCas7-11 processes pre-CRISPR RNA into mature CRISPR RNA (crRNA) and cleaves RNA at positions defined by the target:spacer duplex, without detectable non-specific activity. We engineered Cas7-11 for RNA knockdown and editing in mammalian cells. We show that Cas7-11 has no effects on cell viability, whereas other RNA-targeting tools (such as short hairpin RNAs and Cas13) show substantial cell toxicity4,5. This study illustrates the evolution of a single-protein effector from multisubunit class1 effector complexes, expanding our understanding of the diversity of CRISPR systems. Cas7-11 provides the basis for new programmable RNA-targeting tools that are free of collateral activity and cell toxicity.

With the recent advancements in computer graphics, 3D virtual humans reached to a level of very high appearance realism. They take place in a range of applications such as games, chatbots for customer service and finance, simulations for education and healthcare, remote communication and (social) XR. Yet, their interactivity and movement is limited. We as humans are very receptive of non-verbal behaviors when we engage in social and emotional interactions. In order to interact with virtual humans naturally, 3D digital humans should also be equipped with non-verbal communication skills such as facial expressions, gestures and gaze. As they take place in more interactive applications, the demand to automatically generate their behavior on-the-fly increases. This talk will discuss the role of AI and machine learning algorithms to generate the behavior and movement of virtual humans i.e. using motion capture technology. I will present the state-of-the-art and some of our research on socially interactive virtual humans.

Fifteen years after the invention of Bitcoin, there has been a proliferation of many permissionless blockchains. Each such chain provides a public ledger that can be written to and read from by anyone. In this multi-chain world, a natural question arises: what is the optimal security an existing blockchain, a consumer chain, can extract by only reading and writing to k other existing blockchains, the provider chains In this talk, we will answer this question in three ways: (1) We will first see a protocol, Babylon, where an off-the-shelf proof-of-stake protocol (acting as a consumer chain) checkpoints onto Bitcoin (the provider chain) to reduce its stake withdrawal delay and to resolve issues such as non-slashable long-range safety attacks, low liveness resilience and difficulty to bootstrap from low token valuation. (2) Applying the checkpointing method iteratively, we will then design a protocol called interchain timestamping, which enables a consumer chain to extract the maximum economic security from the provider chains, as quantified by the slashable safety resilience.

Plasmonics and the faithfully related field of metamaterials have been among the most active and exciting areas of photonics over the last two decades. The distinctive ability of plasmonic nanostructures is to condense high amounts of optical energy into sub-wavelength volumes beyond the diffraction limit, enabling linear and nonlinear optical processes. The properties of surface plasmons are strongly related to the materials used and the geometry of the nanostructures. The meticulously designed nanostructures or metamaterials of metals, semiconductors, and 2D materials can enable the tuning of plasmonic resonances ranging from UV to mid-IR. Plasmonic nanostructures can promote a wide range of light-matter interaction processes and be a highly promising candidate for sensing, photovoltaics, nonlinear optics, and photocatalysis.

Complex systems consist of many interconnected components organized as networks. Moreover, the structure of such networks may evolve by their intrinsic rhythms. Our understanding of dynamical complex systems relies on our ability to infer appropriate models from observations. In this talk, we discuss how to learn equations of motion from observational data that underlie evolution and why the recovered dynamics contain higher-order edges, although they are not initially in the model. The results provide a necessary starting point to predict and control the network dynamics.

The adaptability of skeletal muscle mitochondria to various stressors and stimuli is well-known as these organelles are highly dynamic. However, when faced with severe and sustained mitochondrial dysfunction, these mechanisms are insufficient, leading to mitochondrial myopathies that cause significant muscular dysfunction.Mitochondrial aspartyl-tRNA synthetase (DARS2) is an essential enzyme responsible for the initial step of mitochondrial protein synthesis, attaching aspartate to its cognate tRNA. To investigate the impact of DARS2 deficiency on skeletal muscle, we generated a mouse model with muscle-specific deletion of DARS2 (mKO) using the CRE recombinase under the human skeletal muscle-specific