Händelser: Data- och informationsteknikhttp://www.chalmers.se/sv/om-chalmers/kalendariumAktuella händelser på Chalmers tekniska högskolaFri, 09 Mar 2018 08:13:30 +0100http://www.chalmers.se/sv/om-chalmers/kalendariumhttps://www.chalmers.se/sv/institutioner/cse/kalendarium/Sidor/Thesis-Defence-Olof-Mogren.aspxhttps://www.chalmers.se/sv/institutioner/cse/kalendarium/Sidor/Thesis-Defence-Olof-Mogren.aspxOlof Mogren, Data- och informationsteknik<p>MC, lecture hall, Gamla M-huset, Maskin</p><p>Representation learning for natural language</p><br />The advances in artificial intelligence have been astonishing in recent years, with new algorithms showing super-human performance for a wide number of tasks. An important reason for this development is the availability of large datasets and powerful computers, making it possible to train larger machine learning models with higher learning capacity. Artificial neural networks (ANNs) are machine learning models that have been of paramount importance to the development. ANNs are composed of layers of artificial neurons, each of which can only perform a simple computation, but when stacked together in deep architectures, they can be trained to approximate complicated non-linear functions. These models have achieved fantastic results in tasks on various data modalities such as audio, vision, and text. One reason for the success is the internal vector representations computed by the layers, each transforming their input into numerical feature vectors which are increasingly useful for the end task. A complete model is often trained at once (end-to-end learning), and the representations are optimized during training to solve the given task. <br /><br /> This thesis studies the representations computed using artificial neural networks that are trained on and applied to natural language. In paper I and II, we apply learned representations for words to improve the performance of multi-document summarization. In Paper III, we study the use of deep neural sequence models working on the raw character stream as input, and how this class of models can be used to detect medical terms in text (such as drugs, symptoms, and body parts). The system is evaluated on medical health records in Swedish. In paper IV, we propose a novel deep neural sequence model trained to transform words into inflected forms as demonstrated by analogies: “see&quot; is to “sees&quot; as “eat&quot; is to what? The model outperforms previous rule-based approaches by a massive margin, and when inspecting the internal representations computed by this model, one can see that it learns to distinguish classes of transformations of word forms, without being explicitly told to do so. This is an effect from training the model to transform words while provided with the analogous words forms. In other cases, however, the training objective may not provide such cues for the learning algorithm. In Paper V, we study how to improve the way learned representations disentangle the underlying factors of variation in the data. This can be useful for unsupervised representation learning, such as using autoencoders for task agnostic representations or when the final use case is unknown. https://www.chalmers.se/sv/institutioner/cse/kalendarium/Sidor/Licentiate-seminar-Rebekka-Wohlrab.aspxhttps://www.chalmers.se/sv/institutioner/cse/kalendarium/Sidor/Licentiate-seminar-Rebekka-Wohlrab.aspxRebekka Wohlrab, Data- och informationsteknik<p>Jupiter 520, conference room, Jupiter, Campus Lindholmen</p><p>​Continuous Management of Artifacts and Traceability in Large-Scale Agile Systems Engineering</p><br />Context: In large-scale systems engineering, a large number of requirements, models, and other artifacts is created, used, and maintained by many stakeholders. When adopting agile methods and trying to reduce unnecessary documentation, companies need to reevaluate how to manage both artifacts and traceability between them. <br /><br />Objective: The goal of this research is to empirically study and support the continuous management of artifacts and traceability in large-scale agile systems engineering. To get an understanding of the context, we examine challenges with the management of artifacts and traceability. Moreover, we investigate what interrelations exist between the organizational, process, and cultural contexts and the management of artifacts and traceability. To support practitioners, we suggest guidelines and strategies to mitigate challenges. <br /><br />Method: We conducted our research in close collaboration with industry by conducting empirical studies. To gain in-depth insights into the topic and its context, we relied on case studies as well as on an ethnographic study. We used a design-science approach to develop and evaluate applicable guidelines in several iterations. <br /><br />Findings and Conclusions: We found that in large-scale agile systems engineering, the alignment of cross-functional teams is difficult. It is unclear what artifacts are needed and how they should be managed. Our findings suggest that there are two sets of artifacts: Boundary objects (used to communicate across team boundaries) and locally relevant artifacts (used within one team). These artifacts differ in importance and require different management processes. It is beneficial to manage boundary objects in communities of representatives from several teams, and ensure that traceability to locally relevant artifacts is established. We found that with adequate support, traceability can facilitate collaboration across team boundaries and bring new incentives to invest in trace link quality. <br /><br />Future Work: In the future, we plan to tailor the provided guidelines to concrete contexts and put them into practice for specific boundary objects (e.g., in the area of systems architecture). We also plan to explore the use of artifacts and traceability for operations activities, e.g., continuous use, trust, and run-time monitoring.https://www.chalmers.se/sv/om-chalmers/kalendarium/Sidor/Chalmers-Hallbarhetsdag2018.aspxhttps://www.chalmers.se/sv/om-chalmers/kalendarium/Sidor/Chalmers-Hallbarhetsdag2018.aspxChalmers Hållbarhetsdag 2018<p>TBA, Chalmers Johanneberg och Lindholmen</p><p>​Chalmers Hållbarhetsdag är tillbaka – tisdag 23 oktober 2018. Temat för i år är God hälsa och välbefinnande.</p>​<br />Mer information kommer, håll dig uppdaterad genom eventets webbsida: <a href="/sv/om-chalmers/miljo-och-hallbar-utveckling/hallbarhetsdagen2018/Sidor/default.aspx">Chalmers Hållbarhetsdag 2018</a>.<br /><br />Markera dagen i din kalender!<span class="text-normal page-content"></span><br />