This event has a video call. Join: https://meet.google.com/ttt-uona-jmx

Location: Engineering building (1103), room 329 Title:The Case for Reasoning Beyond RecognitionAbstract:Algorithms that can jointly process modalities like images+text are needed for next generation search, accessibility, and robot interaction tools. Simply recognizing objects in images, however, is rarely sufficient; to truly be useful, machines must be capable of deeper commonsense inferences about sophisticated multimodal contexts. I'll ... Read more

Location: Zoom meeting: https://us02web.zoom.us/j/4685913265Title:Interactive Learning with Discriminative Feature Feedback  Abstract:In this talk I will discuss a model of learning with feature-based explanations, that we call Discriminative Feature Feedback. This model formalizes a natural notion of interactive learning with explanations. We study algorithms for this model, including robust algorithms for adversarial and stochastic settings, and derive insightful new results ... Read more

NameLab HeadTitle12:00-12:15Ori ErnstIdo DaganBeyond End-to-End: The Case of Multi Document Summarization12:15-12:30Shon OtmazginYoav GoldbergLingMess & F-COREF: Fast, Accurate, and Easy to Use models for Coreference Resolution12:30-12:45Lior Frenkel Jacob GCalibration of Medical Imaging Classification Systems with Weight Scaling12:45-13:00Coby PensoEthan FetayaFunctional Ensemble Distillation13:00-14:00Lunch

Location:Engineering building (1103), room 329Title:Notions of simplicity in deep learning: From time series to images Abstract:It is standard practice indeep learning to train large models on relatively small datasets. This canpotentially lead to severe overfitting, but more often than not, test error isactually good. This phenomenon has prompted research on the so-called "ImplicitBias of Deep Learning Algorithms". ... Read more

Location:Engineering building (1103), room 329Title:Exploring Deep Neural Collapse via Extended and Controlled Unconstrained Features ModelsAbstract:Training deep neural networks for classification often includes minimizing the training loss beyond the zero training error point. In this phase of training, a "neural collapse" (NC) behavior has been empirically observed: the variability of features (outputs of the penultimate layer) ... Read more

Zoom link: https://us02web.zoom.us/j/4685913265Title:Graph Representation Learning Through Recoverability Abstract:Self-supervised learning methods became popular for graph representation learning because they do not rely on manual labels and offer better generalization. Contrastive methods based on mutual information maximization between augmented instances of the same object are widely used in self-supervised learning of representations. For graph-structured data, however, there are two ... Read more

Zoom link: https://us02web.zoom.us/j/4685913265Title:Toward Fast and Efficient Deep Learning Abstract:Deep Neural Networks (DNNs) are now irreplaceable in various applications. However, DNNs require a vast amount of computational resources. In most cases, complex DNNs training requires several machines working in parallel (most commonly using data parallelism). Moreover, DNNs deployment on devices with limited computational power can be challenging and ... Read more

Location:Engineering building (1103), room 329Title:Exploring the Successes and Limitations of Deep Learning Abstract:In this talk, we will explore the successes and limitations of deep learning networks and highlight the need for more rigorous evaluation. Tree ensemble models often outperform deep learning models for tabular data. However, recently there have been claims that some new deep-learning models ... Read more

Location:Engineering building (1103), room 329Title:Incorporating Commonsense Reasoning into NLP ModelsAbstract:Human language is often ambiguous, underspecified, and grounded in social norms. We employ commonsense knowledge and reasoning abilities to understand others. Endowing NLP models with the same abilities is imperative for reaching human-level language understanding and generation skills. In this talk, I will present several lines ... Read more

Location:Engineering building (1103), room 329 Title:Retrieval in the age of large language modelsAbstract:Large language models have revolutionized natural language processing. However, such models have a limited receptive field and cannot be applied directly on tasks that require processing entire corpora. In this talk, I will talk about the role of information retrieval in the new landscape ... Read more

On Sunday 15.01.23, at 12:00 PM, we will have our second session of students’ presentations. In this session, four students from BIU will present their work. Note that, unlike regular learning club meetings, this meeting will last 2 hours, and will include lunch. It will take place at the Engineering building (1103) in room 329. ... Read more

Location:Engineering building (1103), room 329Title:From theory to practice and back – Stance Detection as a case studyAbstract:Stance detection is an important task, supporting many downstream tasks such as discourse parsing and modeling the propagation of fake news, rumors, and science denial. In this talk we describe a novel framework for stance detection. Our framework is ... Read more