Detection of both biological and chemical environmental toxicants is essential in the assessment of risk to human health. Cell-based biosensors are capable of activity- based detection of toxicity. Chromatophore cells, responsible for the pigmentation of poikilothermic animal, have shown immense potential as cell-based biosensors in the detection of a broad range of environmental toxicants. Chromatophore cells possess the motile pigment granules that intracellularly aggregate or disperse in response to external stimuli. Previous studies have assessed chromatophore cells isolated from red Betta splendens and grey Oncorhynchus tschawytscha fish for use as a biosensor. The objective of this study was to describe blue B. splendens chromatophore cells in tissue culture. Blue B. splendens chromatophore cells were assessed for their longevity in tissue culture and their responses to previously established control agents. Blue B. splendens chromatophore cells were exposed to select chemicals and pathogenic bacteria to assess their ability to respond to environmental toxicants. Three concentrations of mercuric chloride, methyl mercuric chloride, paraquat, sodium arsenite, sodium cyanide chemicals were tested. Bacillus cereus, Bacillus subtilis, Salmonella enterica serovar Enteritidis, and Salmonella enterica serovar Typhimurium were tested. Red B. splendens chromatophore cells were subjected to the select chemical and bacterial toxicants, and observed for their responses. The data collected in this and previous studies were compiled to compare chromatophore cell responses to a broad range of environmental toxicants. Chromatophore cells isolated from both blue and red B. splendens were responsive to methyl mercuric chloride and sodium arsenite. Grey O. tschawytscha chromatophore cells have shown responsiveness to mercuric chloride and sodium arsenite. Blue and red B. splendens chromatophore cells were both responsive to B. cereus and both Salmonella serovars. Grey O. tschawytscha have previously been shown to respond to B. cereus as well. In conclusion, this study reports the chromatophore cells isolated from blue B. splendens in tissue culture and showed similar responsiveness to the selected chemical and bacterial environmental toxicants as chromatophore cells isolated from red and grey colored fish. This study provides compelling evidence that the chromatophore response is not dependent on fish color and that chromatophore cells used for a cell-based detection system may be isolated from different colored fish. / Graduation date: 2012

Background: Multidirectional interactions in social networks can have a profound effect on mate choice behavior; e.g., Poecilia mexicana males show weaker expression of mating preferences when being observed by a rival. This may be an adaptation to reduce sperm competition risk, which arises because commonly preferred female phenotypes will receive attention also from surrounding males, and/or because other males can copy the focal male's mate choice. Do P. mexicana males indeed respond to perceived sperm competition risk? We gave males a choice between two females and repeated the tests under one of the following conditions: (1) an empty transparent cylinder was presented (control); (2) another ("audience") male inside the cylinder observed the focal male throughout the 2nd part, or (3) the audience male was presented only before the tests, but could not eavesdrop during the actual choice tests (non-specific sperm competition risk treatments); (4) the focal male could see a rival male interact sexually with the previously preferred, or (5) with the non-preferred female before the 2nd part of the tests (specific sperm competition risk treatments). Results: The strength of individual male preferences declined slightly also during the control treatment (1). However, this decrease was more than two-fold stronger in audience treatment (2), i.e., with non-specific sperm competition risk including the possibility for visual eavesdropping by the audience male. No audience effect was found in treatments (3) and (5), but a weak effect was also observed when the focal male had seen the previously preferred female sexually interact with a rival male (treatment 4; specific sperm competition risk). Conclusion: When comparing the two 'non-specific sperm competition risk' treatments, a very strong effect was found only when the audience male could actually observe the focal male during mate choice [treatment (2)]. This suggests that focal males indeed attempt to conceal their mating preferences so as to prevent surrounding males from copying their mate choice. When there is no potential for eavesdropping [treatment (3)], non-specific specific sperm competition risk seems to play a minor or no role. Our results also show that P. mexicana males tend to share their mating effort more equally among females when the resource value of their previously preferred mate decreases after mating with a rival male (perceived specific sperm competition risk), but this effect is comparatively weak.

abstract: Computational visual aesthetics has recently become an active research area. Existing state-of-art methods formulate this as a binary classification task where a given image is predicted to be beautiful or not. In many applications such as image retrieval and enhancement, it is more important to rank images based on their aesthetic quality instead of binary-categorizing them. Furthermore, in such applications, it may be possible that all images belong to the same category. Hence determining the aesthetic ranking of the images is more appropriate. To this end, a novel problem of ranking images with respect to their aesthetic quality is formulated in this work. A new data-set of image pairs with relative labels is constructed by carefully selecting images from the popular AVA data-set. Unlike in aesthetics classification, there is no single threshold which would determine the ranking order of the images across the entire data-set. This problem is attempted using a deep neural network based approach that is trained on image pairs by incorporating principles from relative learning. Results show that such relative training procedure allows the network to rank the images with a higher accuracy than a state-of-art network trained on the same set of images using binary labels. Further analyzing the results show that training a model using the image pairs learnt better aesthetic features than training on same number of individual binary labelled images. Additionally, an attempt is made at enhancing the performance of the system by incorporating saliency related information. Given an image, humans might fixate their vision on particular parts of the image, which they might be subconsciously intrigued to. I therefore tried to utilize the saliency information both stand-alone as well as in combination with the global and local aesthetic features by performing two separate sets of experiments. In both the cases, a standard saliency model is chosen and the generated saliency maps are convoluted with the images prior to passing them to the network, thus giving higher importance to the salient regions as compared to the remaining. Thus generated saliency-images are either used independently or along with the global and the local features to train the network. Empirical results show that the saliency related aesthetic features might already be learnt by the network as a sub-set of the global features from automatic feature extraction, thus proving the redundancy of the additional saliency module. / Dissertation/Thesis / Masters Thesis Computer Science 2016

Glioblastoma Multiforme (GBM) is a very aggressive brain tumour. Previous studies have suggested that the morphological distribution of single GBM cells may hold information about the severity. This study aims to find if there is a potential for automated morphological qualification and clustering of GBM cells and what it shows. In this context, phase-contrast images from 10 different GBMcell cultures were analyzed. To test the hypothesis that morphological differences exist between the cell cultures, images of single GBM cells images were created from an image over the well using CellProfiler and Python. Singlecellimages were passed through multiple different feature extraction models to identify the model showing the most promise for this dataset. The features were then clustered and quantified to see if any differentiation exists between the cell cultures. The results suggest morphological feature differences exist between GBM cell cultures when using automated models. The siamese network managed to construct clusters of cells having very similar morphology. I conclude that the 10 cell cultures seem to have cells with morphological differences. This highlights the importance of future studies to find what these morphological differences imply for the patients' survivability and choice of treatment.

Gaze tracking has already shown to be a popular technology for desktop devices. When it comes to gaze tracking for mobile devices, however, there is still a lot of progress to be made. There’s still no high accuracy gaze tracking available that works in an unconstrained setting for mobile devices. This work makes contributions in the area of appearance-based unconstrained gaze estimation. Artificial neural networks are trained on GazeCapture, a publicly available dataset for mobile gaze estimation containing over 2 million face images and corresponding gaze labels. In this work, Siamese neural networks are trained to learn linear distances between face images for different gaze points. Then, during inference, calibration points are used to estimate gaze points. This approach is shown to be an effective way of utilizing calibration points in order to improve the result of gaze estimation. / Ögonblickspårning har redan etablerat sig som en populär teknologi för stationära enheter. När det dock gäller mobila enheter så finns det framsteg att göra. Det saknas fortfarande en lösning för ögonblickspårning som fungerar i en undantagsfri miljö för mobila enheter. Detta examensarbete ämnar att bidra till en sådan lösning. Artificiella neurala nätverk tränas på GazeCapture, en allmänt tillgänglig datasamling som består av över 2 miljoner ansiktsbilder samt korresponderande etikett för ögonblickspunkt. I detta examensarbete tränas Siamesiska neurala nätverk för att lära sig det linjära avståndet mellan två ögonblickspunkter. Sedan utnyttjas en samling med kalibreringsbilder för att estimera ögonblickspunkter. Denna teknik visar sig vara ett effektivt sätt att nyttja kalibreringsbilder med målet att förbättra resultatet för ögonblicksestimering.

In the game League of Legends (LoL) there are several different regions globally withtheir own rating distribution. The purpose of this thesis is to examine if there are anydifferences in playing strength between the regions, and if so quantify what the offsetsare numerically.Data of matches played online are available publicly. We extracted 8.7 million matchesin total and over 600 different features per match. Each match is also annotated with alocal rating - which represents the rank it was played at. All these matches are betweenteams from similar regions and not across regions - hence the rating is a local one andnot a global one. Absence of a global score prevents us from comparing matches acrossregions. Our goal is to rank the different regions by developing a model that can predicta global score using the data available for local ratings.We first develop a Deep Neural Network (DNN) which is trained on equal amounts ofdata from all the regions to predict a global rating. We then use a Siamese Neural Network (SNN), with the purpose of generating a distribution that would be comparableto the true distribution of ratings. In both the above experiments we hide the regioninformation from the network. We also developed a model that is provided region information in a separate layer while training. The outcome of the DNN model is validatedby using the outcomes of SNN and region-aware models. In order to further improvethe results, we normalize the data with respect to the duration of a match. We performfurther experiments where a model is trained on matches from one specific region andthen use it for predicting ratings of matches from other regions.The results allowed us to rank the different regions based on their performance. Someof the results were surprising - for instance the experiments suggests that Japan andOceania, who has very little presence on the professional e-sports scene, are in the top.

Animals coordinate their behavior with each other during cooperative and agonistic social interactions. Such coordination often adopts the form of “turn-taking”, in which the interactive partners alternate the performance of a behavior. Apart from acoustic communication, how turn taking is coordinated, is not well known. Furthermore, the neural substrates that regulate persistence in engaging in social interactions are poorly studied. Here, we use Siamese fighting fish (Betta splendens), to study visually-driven turn-taking aggressive behavior. Using encounters with real conspecifics and with computer animations, we discover the visual cues from an opponent and the behavioral dynamics that generate turn taking. Through a brain-wide screen of neuronal activity during aggressive behavior, followed by targeted brain lesions, we then discover that the caudal portion of the dorsomedial telencephalon, an amygdala-like region, promotes continuous participation in aggressive interactions. Our work highlights how dynamic visual cues shape the rhythm of social interactions at multiple timescales and points to the pallial amygdala as a region controlling the drive to engage in such interactions.

Surface topography systems enable the capture of spinal dynamic movement. A visualization of possible unique movement patterns appears to be difficult due to large intraclass and small inter-class variabilities. Therefore, we investigated a visualization approach using Siamese neural networks (SNN) and checked, if the identification of individuals is possible based on dynamic spinal data. The presented visualization approach seems promising in visualizing subjects in the presence of intraindividual variability between different gait cycles as well as day-to-day variability. Overall, the results indicate a possible existence of a personal spinal ‘fingerprint’. The work forms the basis for an objective comparison of subjects and the transfer of the method to clinical use cases.

Unsupervised learning of speech is concerned with automatically finding patterns such as words or speech sounds, without supervision in the form of orthographical transcriptions or a priori knowledge of the language. However, a fundamental problem is that unsupervised speech learning methods tend to discover highly speaker-specific and context-dependent representations of speech. We propose a method for improving the quality of posteriorgrams generated from an unsupervised model through partitioning of the latent classes discovered by the model. We do this by training a sparse siamese model to find a linear transformation of input posteriorgrams, extracted from the unsupervised model, to lower-dimensional posteriorgrams. The siamese model makes use of same-category and different-category speech fragment pairs obtained through unsupervised term discovery. After training, the model is converted into an exact partitioning of the posteriorgrams. We evaluate the model on the minimal-pair ABX task in the context of the Zero Resource Speech Challenge. We are able to demonstrate that our method significantly reduces the dimensionality of standard Gaussian mixture model posteriorgrams, while also making them more speaker invariant. This suggests that the model may be viable as a general post-processing step to improve probabilistic acoustic features obtained by unsupervised learning. / Obevakad inlärning av tal innebär att automatiskt hitta mönster i tal, t ex ord eller talljud, utan bevakning i form av ortografiska transkriptioner eller tidigare kunskap om språket. Ett grundläggande problem är dock att obevakad talinlärning tenderar att hitta väldigt talar- och kontextspecifika representationer av tal. Vi föreslår en metod för att förbättra kvaliteten av posteriorgram genererade med en obevakad modell, genom att partitionera de latenta klasserna funna av modellen. Vi gör detta genom att träna en gles siamesisk modell för att hitta en linjär transformering av de givna posteriorgrammen, extraherade från den obevakade modellen, till lågdimensionella posteriorgram. Den siamesiska modellen använder sig av talfragmentpar funna med obevakad ordupptäckning, där varje par består av fragment som antingen tillhör samma eller olika klasser. Den färdigtränade modellen görs sedan om till en exakt partitionering av posteriorgrammen. Vi följer Zero Resource Speech Challenge, och evaluerar modellen med hjälp av minimala ordpar-ABX-uppgiften. Vi demonstrerar att vår metod avsevärt minskar posteriorgrammens dimensionalitet, samtidigt som posteriorgrammen blir mer talarinvarianta. Detta antyder att modellen kan vara användbar som ett generellt extra steg för att förbättra probabilistiska akustiska särdrag från obevakade modeller.

Product matching in e-commerce is an area that faces more and more challenges with growth in the e-commerce marketplace as well as variation in the quality of data available online for each product. Product matching for e-commerce provides competitive possibilities for vendors and flexibility for customers by identifying identical products from different sources. Traditional methods in product matching are often conducted through rule-based methods and methods tackling the issue through machine learning usually do so through unimodal systems. Moreover, existing methods would tackle the issue through product identifiers which are not always unified for each product. This thesis provides multimodal approaches through product name, description, and image to the problem area of product matching that outperforms unimodal approaches. Three multimodal approaches were taken, one unsupervised and two supervised. The unsupervised approach uses straight-forward embedding space to nearest neighbor search that provides better results than unimodal approaches. One of the supervised multimodal approaches uses Siamese network on the embedding space which outperforms the unsupervised multi- modal approach. Finally, the last supervised approach instead tackles the issue by exploiting distance differences in each modality through logistic regression and a decision system that provided the best results. / Produktmatchning inom e-handel är ett område som möter fler och fler utmaningar med hänsyn till den tillväxt som e-handelsmarknaden undergått och fortfarande undergår samt variation i kvaliteten på den data som finns tillgänglig online för varje produkt. Produktmatchning inom e-handel är ett område som ger konkurrenskraftiga möjligheter för leverantörer och flexibilitet för kunder genom att identifiera identiska produkter från olika källor. Traditionella metoder för produktmatchning genomfördes oftast genom regelbaserade metoder och metoder som utnyttjar maskininlärning gör det vanligtvis genom unimodala system. Dessutom utnyttjar mestadels av befintliga metoder produktidentifierare som inte alltid är enhetliga för varje produkt mellan olika källor. Denna studie ger istället förslag till multimodala tillvägagångssätt som istället använder sig av produktnamn, produktbeskrivning och produktbild för produktmatchnings-problem vilket ger bättre resultat än unimodala metoder. Tre multimodala tillvägagångssätt togs, en unsupervised och två supervised. Den unsupervised metoden använder embeddings vektorerna rakt av för att göra en nearest neighborsökning vilket gav bättre resultat än unimodala tillvägagångssätt. Ena supervised multimodal tillvägagångssätten använder siamesiska nätverk på embedding utrymmet vilket gav resultat som överträffade den unsupervised multimodala tillvägagångssättet. Slutligen tar den sista supervised metoden istället avståndsskillnader i varje modalitet genom logistisk regression och ett beslutssystem som gav bästa resultaten.