Biblio

We propose Cleann, the first end-to-end framework that enables online mitigation of Trojans for embedded Deep Neural Network (DNN) applications. A Trojan attack works by injecting a backdoor in the DNN while training; during inference, the Trojan can be activated by the specific backdoor trigger. What differentiates Cleann from the prior work is its lightweight methodology which recovers the ground-truth class of Trojan samples without the need for labeled data, model retraining, or prior assumptions on the trigger or the attack. We leverage dictionary learning and sparse approximation to characterize the statistical behavior of benign data and identify Trojan triggers. Cleann is devised based on algorithm/hardware co-design and is equipped with specialized hardware to enable efficient real-time execution on resource-constrained embedded platforms. Proof of concept evaluations on Cleann for the state-of-the-art Neural Trojan attacks on visual benchmarks demonstrate its competitive advantage in terms of attack resiliency and execution overhead.

An approximate sparse recovery system in ℓ1 norm consists of parameters k, ε, N; an m-by-N measurement Φ; and a recovery algorithm R. Given a vector, x, the system approximates x by &xwidehat; = R(Φ x), which must satisfy ‖ &xwidehat;-x‖1 ≤ (1+ε)‖ x - xk‖1. We consider the “for all” model, in which a single matrix Φ, possibly “constructed” non-explicitly using the probabilistic method, is used for all signals x. The best existing sublinear algorithm by Porat and Strauss [2012] uses O(ε−3klog (N/k)) measurements and runs in time O(k1 − αNα) for any constant α textgreater 0. In this article, we improve the number of measurements to O(ε − 2klog (N/k)), matching the best existing upper bound (attained by super-linear algorithms), and the runtime to O(k1+βpoly(log N,1/ε)), with a modest restriction that k ⩽ N1 − α and ε ⩽ (log k/log N)γ for any constants α, β, γ textgreater 0. When k ⩽ log cN for some c textgreater 0, the runtime is reduced to O(kpoly(N,1/ε)). With no restrictions on ε, we have an approximation recovery system with m = O(k/εlog (N/k)((log N/log k)γ + 1/ε)) measurements. The overall architecture of this algorithm is similar to that of Porat and Strauss [2012] in that we repeatedly use a weak recovery system (with varying parameters) to obtain a top-level recovery algorithm. The weak recovery system consists of a two-layer hashing procedure (or with two unbalanced expanders for a deterministic algorithm). The algorithmic innovation is a novel encoding procedure that is reminiscent of network coding and that reflects the structure of the hashing stages. The idea is to encode the signal position index i by associating it with a unique message mi, which will be encoded to a longer message m′i (in contrast to Porat and Strauss [2012] in which the encoding is simply the identity). Portions of the message m′i correspond to repetitions of the hashing, and we use a regular expander graph to encode the linkages among these portions. The decoding or recovery algorithm consists of recovering the portions of the longer messages m′i and then decoding to the original messages mi, all the while ensuring that corruptions can be detected and/or corrected. The recovery algorithm is similar to list recovery introduced in Indyk et al. [2010] and used in Gilbert et al. [2013]. In our algorithm, the messages \mi\ are independent of the hashing, which enables us to obtain a better result.

In this work, a new fingerprinting-based localization algorithm is proposed for an underwater medium by utilizing ultra-wideband (UWB) signals. In many conventional underwater systems, localization is accomplished by utilizing acoustic waves. On the other hand, electromagnetic waves haven't been employed for underwater localization due to the high attenuation of the signal in water. However, it is possible to use UWB signals for short-range underwater localization. In this work, the feasibility of performing localization for an underwater medium is illustrated by utilizing a fingerprinting-based localization approach. By employing the concept of compressive sampling, we propose a sparsity-based localization method for which we define a system model exploiting the spatial sparsity.