We use cookies to analyze the browsing and usage of our website and to personalize your experience. You can disable these technologies at any time, but this may limit certain functionalities of the site. Read our Privacy Policy for more information.
Setting cookies
You can enable and disable the types of cookies you wish to accept. However certain choices you make could affect the services offered on our sites (e.g. suggestions, personalised ads, etc.).
Essential cookies
These cookies are necessary for the operation of the site and cannot be deactivated. (Still active)
Analytics cookies
Do you accept the use of cookies to measure the audience of our sites?
Multimedia Player
Do you accept the use of cookies to display and allow you to watch the video content hosted by our partners (YouTube, etc.)?
Publications
A Probabilistic Framework for Mutation Testing in Deep Neural Networks
Background Reward processing has been proposed to underpin the atypical social feature of autism spectrum disorder (ASD). However, previous … (see more)neuroimaging studies have yielded inconsistent results regarding the specificity of atypicalities for social reward processing in ASD. Aims Utilising a large sample, we aimed to assess reward processing in response to reward type (social, monetary) and reward phase (anticipation, delivery) in ASD. Method Functional magnetic resonance imaging during social and monetary reward anticipation and delivery was performed in 212 individuals with ASD (7.6–30.6 years of age) and 181 typically developing participants (7.6–30.8 years of age). Results Across social and monetary reward anticipation, whole-brain analyses showed hypoactivation of the right ventral striatum in participants with ASD compared with typically developing participants. Further, region of interest analysis across both reward types yielded ASD-related hypoactivation in both the left and right ventral striatum. Across delivery of social and monetary reward, hyperactivation of the ventral striatum in individuals with ASD did not survive correction for multiple comparisons. Dimensional analyses of autism and attention-deficit hyperactivity disorder (ADHD) scores were not significant. In categorical analyses, post hoc comparisons showed that ASD effects were most pronounced in participants with ASD without co-occurring ADHD. Conclusions Our results do not support current theories linking atypical social interaction in ASD to specific alterations in social reward processing. Instead, they point towards a generalised hypoactivity of ventral striatum in ASD during anticipation of both social and monetary rewards. We suggest this indicates attenuated reward seeking in ASD independent of social content and that elevated ADHD symptoms may attenuate altered reward seeking in ASD.
In this article, we consider the problem of controlling an unknown linear quadratic Gaussian (LQG) system consisting of multiple subsystems … (see more)connected over a network. Our goal is to minimize and quantify the regret (i.e., loss in performance) of our learning and control strategy with respect to an oracle who knows the system model. Upfront viewing the interconnected subsystems globally and directly using existing LQG learning algorithms for the global system results in a regret that increases super-linearly with the number of subsystems. Instead, we propose a new Thompson sampling-based learning algorithm which exploits the structure of the underlying network. We show that the expected regret of the proposed algorithm is bounded by
2023-03-01
IEEE Transactions on Control of Network Systems (published)
With the increasing adoption of digital health platforms through mobile apps and online services, people have greater flexibility connecting… (see more) with medical practitioners, pharmacists, and laboratories and accessing resources to manage their own health-related concerns. Many healthcare institutions are connecting with each other to facilitate the exchange of healthcare data, with the goal of effective healthcare data management. The contents generated over these platforms are often shared with third parties for a variety of purposes. However, sharing healthcare data comes with the potential risk of exposing patients’ sensitive information to privacy threats. In this article, we address the challenge of sharing healthcare data while protecting patients’ privacy. We first model a complex healthcare dataset using a heterogeneous information network that consists of multi-type entities and their relationships. We then propose DiffHetNet, an edge-based differentially private algorithm, to protect the sensitive links of patients from inbound and outbound attacks in the heterogeneous health network. We evaluate the performance of our proposed method in terms of information utility and efficiency on different types of real-life datasets that can be modeled as networks. Experimental results suggest that DiffHetNet generally yields less information loss and is significantly more efficient in terms of runtime in comparison with existing network anonymization methods. Furthermore, DiffHetNet is scalable to large network datasets.
2023-02-28
ACM Transactions on Knowledge Discovery from Data (published)
Recent advances in deep learning have relied heavily on the use of large Transformers due to their ability to learn at scale. However, the c… (see more)ore building block of Transformers, the attention operator, exhibits quadratic cost in sequence length, limiting the amount of context accessible. Existing subquadratic methods based on low-rank and sparse approximations need to be combined with dense attention layers to match Transformers, indicating a gap in capability. In this work, we propose Hyena, a subquadratic drop-in replacement for attention constructed by interleaving implicitly parametrized long convolutions and data-controlled gating. In recall and reasoning tasks on sequences of thousands to hundreds of thousands of tokens, Hyena improves accuracy by more than 50 points over operators relying on state-spaces and other implicit and explicit methods, matching attention-based models. We set a new state-of-the-art for dense-attention-free architectures on language modeling in standard datasets (WikiText103 and The Pile), reaching Transformer quality with a 20% reduction in training compute required at sequence length 2K. Hyena operators are twice as fast as highly optimized attention at sequence length 8K, and 100x faster at sequence length 64K.