This post contains a list of the AI-related seminars that are scheduled to take place between 9 May 2022 and 30 June 2022. All events detailed here are free and open for anyone to attend virtually. Note: this event runs for four days – 9-12 May. Instance-adaptive data compression: Improving Neural Codecs by Training on the Test Set Speaker: Ties van Rozendaal Organised by: University of California, Irvine The live stream is here. Kernel-based robust inference for intractable likelihood models Speaker: François-Xavier Briol Organised by: Finnish Centre for AI Zoom link is here.

Automatic depression detection has attracted increasing amount of attention but remains a challenging task. Psychological research suggests that depressive mood is closely related with emotion expression and perception, which motivates the investigation of whether knowledge of emotion recognition can be transferred for depression detection. This paper uses pretrained features extracted from the emotion recognition model for depression detection, further fuses emotion modality with audio and text to form multimodal depression detection. The proposed emotion transfer improves depression detection performance on DAIC-WOZ as well as increases the training stability. The analysis of how the emotion expressed by depressed individuals is further perceived provides clues for further understanding of the relationship between depression and emotion.

Abstract--Artificial Intelligence (AI) has found its applications in a variety of environments ranging from data science to cybersecurity. AI helps break through the limitations of traditional algorithms and provides more efficient and flexible methods for solving problems. In this paper, we focus on the applications of artificial intelligence in authentication, which is used in a wide range of scenarios including facial recognition to access buildings, keystroke dynamics to unlock smartphones. With the emerging AI-assisted authentication schemes, our comprehensive survey provides an overall understanding on a high level, which paves the way for future research in this area. In contrast to other relevant surveys, our research is the first of its kind to focus on the roles of AI in authentication. Learning and neural networks are The traditional password-based authentication method has two main mechanisms used in AI. Learning is the process of slowly faded out due to its inadequate ...

We propose a framework to modularize the training of neural language models that use diverse forms of sentence-external context (including metadata) by eliminating the need to jointly train sentence-external and within-sentence encoders. Our approach, contextual universal embeddings (CUE), trains LMs on one set of context, such as date and author, and adapts to novel metadata types, such as article title, or previous sentence. The model consists of a pretrained neural sentence LM, a BERT-based context encoder, and a masked transformer decoder that estimates LM probabilities using sentence-internal and sentence-external information. When context or metadata are unavailable, our model learns to combine contextual and sentence-internal information using noisy oracle unigram embeddings as a proxy. Real contextual information can be introduced later and used to adapt a small number of parameters that map contextual data into the decoder's embedding space. We validate the CUE framework on a NYTimes text corpus with multiple metadata types, for which the LM perplexity can be lowered from 36.6 to 27.4 by conditioning on context. Bootstrapping a contextual LM with only a subset of the context/metadata during training retains 85\% of the achievable gain. Training the model initially with proxy context retains 67% of the perplexity gain after adapting to real context. Furthermore, we can swap one type of pretrained sentence LM for another without retraining the context encoders, by only adapting the decoder model. Overall, we obtain a modular framework that allows incremental, scalable training of context-enhanced LMs.

Randomized Hadamard Transforms (RHTs) have emerged as a computationally efficient alternative to the use of dense unstructured random matrices across a range of domains in computer science and machine learning. For several applications such as dimensionality reduction and compressed sensing, the theoretical guarantees for methods based on RHTs are comparable to approaches using dense random matrices with i.i.d.\ entries. However, several such applications are in the low-dimensional regime where the number of rows sampled from the matrix is rather small. Prior arguments are not applicable to the high-dimensional regime often found in machine learning applications like kernel approximation. Given an ensemble of RHTs with Gaussian diagonals, $\{M^i\}_{i = 1}^m$, and any $1$-Lipschitz function, $f: \mathbb{R} \to \mathbb{R}$, we prove that the average of $f$ over the entries of $\{M^i v\}_{i = 1}^m$ converges to its expectation uniformly over $\| v \| \leq 1$ at a rate comparable to that obtained from using truly Gaussian matrices. We use our inequality to then derive improved guarantees for two applications in the high-dimensional regime: 1) kernel approximation and 2) distance estimation. For kernel approximation, we prove the first \emph{uniform} approximation guarantees for random features constructed through RHTs lending theoretical justification to their empirical success while for distance estimation, our convergence result implies data structures with improved runtime guarantees over previous work by the authors. We believe our general inequality is likely to find use in other applications.

If faced with the choice of sending a swarm of full-sized, distinct robots to space, or a large crew of smaller robotic modules, you might want to enlist the latter. Modular robots, like those depicted in films such as "Big Hero 6," hold a special type of promise for their self-assembling and reconfiguring abilities. But for all of the ambitious desire for fast, reliable deployment in domains extending to space exploration, search and rescue, and shape-shifting, modular robots built to date are still a little clunky. They're typically built from a menagerie of large, expensive motors to facilitate movement, calling for a much-needed focus on more scalable architectures -- both up in quantity and down in size. Scientists from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) called on electromagnetism -- electromagnetic fields generated by the movement of electric current -- to avoid the usual stuffing of bulky and expensive actuators into individual blocks.

Scientists from MIT's Computer Science and Artificial Intelligence Laboratory ( CSAIL) and the University of Calgary have developed a modular robot system that can morph into different shapes. ElectroVoxels don't have any motors or moving parts. Instead, they use electromagnets to shift around each other. Each edge of an ElectroVoxel cube is an electromagnetic ferrite core wrapped with copper wire. The length of each ElectroVoxel side is around 60 millimeters.

AltaML was founded in 2018, employs around 130 employees and has offices in Edmonton, Calgary and Toronto. The company works in industries such as oil and gas, banking, forestry, agriculture and health. "We work with them to uncover those possibilities for the application of machine learning," says Rabelo. "When we identify those opportunities, we develop studies – basically, experiments – to see if our hypotheses really hold true when we apply them to real world data." "As we validate those hypotheses, those opportunities move on a chain and eventually they reach solution phase, where they are deployed to production. They are developed as part of software system or an [application programming interface] or something that can be directly deployed to those industries, to those clients."

Unlike the classical linear model, nonlinear generative models have been addressed sparsely in the literature. This work aims to bring attention to these models and their secrecy potential. To this end, we invoke the replica method to derive the asymptotic normalized cross entropy in an inverse probability problem whose generative model is described by a Gaussian random field with a generic covariance function. Our derivations further demonstrate the asymptotic statistical decoupling of Bayesian inference algorithms and specify the decoupled setting for a given nonlinear model. The replica solution depicts that strictly nonlinear models establish an all-or-nothing phase transition: There exists a critical load at which the optimal Bayesian inference changes from being perfect to an uncorrelated learning. This finding leads to design of a new secure coding scheme which achieves the secrecy capacity of the wiretap channel. The proposed coding has a significantly smaller codebook size compared to the random coding scheme of Wyner. This interesting result implies that strictly nonlinear generative models are perfectly secured without any secure coding. We justify this latter statement through the analysis of an illustrative model for perfectly secure and reliable inference.

The University of Calgary acknowledges the traditional territories of the people of the Treaty 7 region in Southern Alberta, which includes the Blackfoot Confederacy (comprised of the Siksika, Piikani, and Kainai First Nations), as well as the Tsuut'ina First Nation, and the Stoney Nakoda (including the Chiniki, Bearspaw and Wesley First Nations). The City of Calgary is also home to Metis Nation of Alberta, Region 3. The University of Calgary acknowledges the impact of colonization on Indigenous peoples in Canada and is committed to our collective journey towards reconciliation to create a welcome and inclusive campus that encourages Indigenous ways of knowing, doing, connecting and being.