Events

This talk is inspired by the role of environments in shaping behaviors. Through the discussion of two studies, we will ask: how do/should we construct environments in ways that regulate problematic behaviors We will first focus on our social environments and their role in the misinformation ecosystem.

Information theory is rooted in two fundamental principles: redundancy and the distance between codewords. The classical Hamming metric measures the distance between codewords by counting the number of positions where they differ. In recent decades, information theorists have explored the use of ordered structures to gain deeper insights into coding theory. A prominent example is the Niederreiter-Tsfasman-Rozenbloom (NTR) metric, defined on matrices. In this talk, we will discuss our variants of the NTR metric and their applications in the setting of algebraic geometry codes.

Nonlinear dispersive equations arise as reduced mathematical models from governing equations of mathematical physics, such as the Navier-Stokes and Maxwell equations. These reduced models combine the leading-order balance between nonlinear and dispersive effects present in wave propagation. The existence and stability of coherent structures such as traveling, standing, or periodic wave solutions, and the long-time dynamics near these coherent structures are of great importance. In this talk, I will present some of the ideas (ODE and PDE theory) that one typically employs to define and study the stability of such structures. I will also discuss the models I have investigated for the existence and dynamics of nonlinear waves.

Machine learning and generative AI models are transforming all aspects of human experience. In the scientific domain, machine learning models are finding application areas in solving complex problems such as protein folding, chemical property prediction and the design of new materials and drugs. In this talk, we will use two recent publications from Microsoft Research AI for Science (MatterSim: https://microsoft.github.io/mattersim/ and MatterGen) to discuss how the emergence of large-scale AI is transforming the sciences, focusing particularly in materials design and discovery.

The clean sky II core partnership SHERLOC project provides a detailed assessment of current Structural Health Monitoring (SHM) technologies, including piezoelectric transducers (PZT), fiber Bragg gratings (FBG), and magnetostrictive sensors, as well as methodologies for use in composite airframes. These evaluations are tailored to reflect the operational and environmental conditions typical of regional aircraft fleets. The project addresses key challenges in maintaining structural integrity and reliability while aiming to optimize efficiency and reduce maintenance costs. A building-block approach has been developed, aligned with certification guidelines, to evaluate SHM systems as integrated solutions.

The presence of sarcasm in online communication has motivated an increasing number of computational investigations of the phenomenon across the community. In this talk, we first overview an analysis into the socio-demographic ecology of sarcastic exchanges between humans, focusing on how the effectiveness of such exchanges can be influenced by the socio-demographic similarity between the humans. Next, we challenge the motivation of a specific endeavour of the community: mainly, that of augmenting dialogue systems with the ability to generate sarcastic responses. We overview a series of social experiments that provide guidelines for dialogue systems concerning the appropriateness of generating sarcastic responses, and the formulation of such responses. To enable the experiments, we introduce a sarcastic response generator grounded in the implicit display theory.

I will give a brief account of the current state of affairs in a very classical problem which, while going all the way back to Cayley, Solomon, and Schur, has become quite popular in the recent decades. It started from counting/bounding the number of lines on a smooth complex spatial surface. However, even though the original problem was closed long ago (essentially by Segre), the modern approaches and technology let us answer questions that we did not even dare to ask before and,as it is often the case, the more we learn the more we want to know, each result giving rise to a number of conjectures still waiting for their resolution. Thus, we can allow surfaces with singularities, consider other (quasi-)polarizations, count smooth rational curves of other degrees (conics, twisted cubics, etc.), and work over other fields of definition, including non-algebraically closed ones. Concentrating on the particular case of K3-surfaces (spatial quartics and plane sextic curves in the original setting), I will discuss the recent results and the tools used and work out a few examples, both classical and advanced.

With the journey towards 6G, we envision the network evolving into an AI-native platform that not only utilizes AI internally but also exposes its AI capabilities to support a wide range of applications. At its core, this evolution involves the incorporation of AI-driven solutions, elevating network performance to unprecedented levels while introducing an era of intelligent network automation. This shift represents a cornerstone in network architecture evolution towards AI-Nativeness, embodying intelligence at every stage. The AI-native concept involves integrating intrinsic and reliable AI capabilities seamlessly into the design, deployment, operation, and maintenance of networks. Moreover, this paradigm embraces a data-driven, knowledge-based ecosystem to develop novel AI-based functionalities and replace static mechanisms with adaptive AI solutions. Key components include MLOps for real-time decision-making, DataOps for distributed data management, AI-as-a-Service (AIaaS) for democratizing AI capabilities, and Zero-Touch Management for autonomous network operations. Additionally, the exposure of AI services, AIaaS, is one aspect of this overall vision. The idea is that the network, functioning as an AIaaS provider, can offer pre-built AI models, datasets, algorithms, and tools that applications can readily access through Application Programming Interfaces (APIs). Consequently, the primary advantage of AIaaS lies in enabling applications to leverage AI functionalities without having to construct and manage their own AI infrastructure. Therefore, the network transforms into a platform that fosters innovation for a variety of use cases.

The traditional medical approach has been to categorize human disorders into common and rare forms. However, recent knowledge, fueled mainly by advances in high-throughput omics applications, demonstrated that the actual separation is determined by the rarity or commonality - that is the population allele frequency - of the genetic variants underlying susceptibility to or causality of the disease occurrence. As such, we have been investigating the genetic variants causing vascular, developmental, degenerative, and oncological diseases of the nervous system over the last two decades. In this talk, I will be mentioning our approach to gene identification followed by functional characterization of such variants utilizing multi-omics applications, model organisms, and more recently, patient derived in vitro and ex vivo 2- and 3 dimensional cellular systems.

A theoretical understanding of generalization remains an open problem for many machine learning models, including deep networks where overparameterization leads to better performance. Here, we study this problem for kernel regression, which, besides being a popular machine learning method, also describes infinitely overparameterized neural networks. We develop an analytical theory of generalization in kernel regression using the replica method of statistical mechanics. This theory is applicable to any kernel and data distribution. Experiments with practical kernels, including those arising from wide neural networks, show perfect agreement with our theory. Further, our theory accurately predicts the generalization performance of neural networks with modest widths. We provide an in-depth analysis of our analytical expression for kernel generalization. We show that kernel machines employ an inductive bias towards simple functions, preventing them from overfitting the data. We characterize whether a kernel is compatible with a learning task in terms of sample efficiency. We identify a first-order phase transition in our theory where more data may impair generalization when the task is noisy or not expressible by the kernel. Finally, we extend these results to out-of-distribution generalization.

The Inverse Galois problem asks whether every finite group G is the Galois group of some Galois extension of the field of rational numbers Q. The Malle conjecture is a quantitative version of it: it predicts the number of Galois extensions of Q with the Galois group G of bounded size (such as the discriminant). In this talk we will present how one can generalize the conjecture for finite group schemes. We show how the generalization helps explain inconsistencies of the Malle conjecture found by Klüners. The talk is based on a joint work with Takehiko Yasuda.

Distributed systems are prone to concurrency bugs due to the nondeterminism in the interleavings of the concurrent events in an execution. Detecting and diagnosing concurrency bugs in distributed systems is critical since unforeseen interleavings of concurrent messages, network, or process faults can result in unexpected, erroneous system behavior. However, concurrency bugs are hard to detect as they are triggered only in some subtle interleavings of the events. Random testing is a practical way of searching for bugs in large distributed systems. While na

Abiotic stresses, such as drought, salinity, extreme temperatures, and heavy metal toxicity, significantly constrain crop productivity and food security worldwide. Seaweed-based biostimulants, largely manufactured using brown and red seaweeds, have emerged as a tool to mitigate abiotic stresses. These biostimulants enhance plant resilience by modulating physiological and molecular mechanisms. The bioactive compounds such as polysaccharides, polyphenols, and secondary metabolites present in seaweed biostimulants affect stress-responsive pathways, enhancing photosynthetic efficiency, water-use efficiency, and cellular homeostasis. At a molecular level, the bioactive molecules affect transcriptional and post-translational processes involved in the regulation of gene expression leading to better plant performance under abiotic stresses. The presentation will include result of studies conducted at the Marine bioproducts laboratory, Dalhousie University.

Cellular therapies using engineered cells are triggering a paradigm shift in the development of biopharmaceuticals. Instead of using simply a single protein (e.g. antibody, human proteins etc), microbial therapeutics enables us to design biological machines with multifunctional properties. Microbial therapeutics are mainly designed as three modules, including: sensing- targeting and delivery. In our laboratory we have been developing microbial therapeutics for various diseases including Type1 diabetes, breast/column cancer, anemia and rare metabolic diseases. We have established various cellular protein secretion systems for the delivery of the biotechnological drugs. In addition to logic gated cell surface systems, recently we have developed a complex genetic circuitry to control the permeability/leakiness of the cell membrane in a probiotic bacterium E. coli Nissle 1917 for payload delivery of protein drugs. I will give an overview of our recent research efforts focused but not limited only with living drugs.