Transfer Learning for Human & AI

Recent research explores incorporating knowledge graphs (KG) into e-commerce recommendation systems . This can be achieved by explicit KG reasoning, where a model starts from a usernode, sequentially determines the next step, and walks towards an item node ofpotential interest to the user .…

Centrality measures characterize important nodes in networks . Efficiently computing such nodes has received a lot of attention . When considering thegeneralization of computing central groups of nodes, challenging optimizationproblems occur . In this work, we study two such problems, group-harmonicmaximization and group-closeness maximization .…

Semantically-aligned $(speech, image)$ datasets can be used to explore”visually-grounded speech” Previous results have tended to show low rates of recall in $speech\rightarrow image$ and $image \rightarrow speech$ queries . Choosing appropriate neural architectures for encoders in the speech andimage branches and using large datasets, one can obtain competitive recall rates without any reliance on any pre-trained initialization or featureextraction .…

If a linear code is invariant under the action of a doubly transitive permutation group, it achieves the capacity of erasurechannel . Therefore, it is of sufficient interest to classify all codes,invariant under such permutation groups . We take a step in this direction and give a review of all suitable groups and the known results on codes .…

Orthogonal time frequency and space (OTFS) modulation is a promising technology that satisfies high Doppler requirements for future mobile systems . The proposed estimation provides new insight into the OTFS input-output relation inthe DD domain as a 2D circular convolution with a small approximation .…

Machine Learning Challenge was organized for the EuropeanSpace Agency’s upcoming Ariel mission . Successful solutions either construct highlynon-linear (w.r.t. the raw data) models with minimal preprocessing -deep neuralnetworks and ensemble methods- or amount to obtaining meaningful statistics from the light curves, constructing linear models on which yields comparably good predictive performance.…

This work defines the paper length prediction task as are the problem and reports several experimental results using popular machine learning models . We also create a huge dataset of publication metadata and the respective lengths in number of pages .…

The lock-free, ordered, linked list is an important, standard example of aconcurrent data structure . Failed compare-and-swap operations lead toretraversal of the entire list . We alleviate this drawback by maintaining approximate backwards pointers that are used to find a closer starting position in the list for operation retry .…

The proposed method has the guarantee that the discrepancy between the pruned network and the original network decays with exponentiallyfast rate w.r.t. the size of the prune network, under weak assumptions that apply for most practical settings . Empirically, our method improves prior artson pruning various network architectures including ResNet, MobilenetV2/V3 onImageNet .…

Reconfigurable intelligent surface (RIS) has the potential to significantlyenhance the network secure transmission performance . However, due to the passive nature ofeavesdroppers and the cascaded channel brought by the RIS, the eavesdroppers’ channel state information is imperfectly obtained at the base station .…

We argue that an excess in entanglement between the visible and hidden units in a Quantum Neural Network can hinder learning . We show that for any bounded objectivefunction on the visible layers, the Lipshitz constants of the expectation value of that objective function will scale inversely with the dimension of the hidden-subsystem with high probability.…

The remarkable success of pretrained language models has motivated the study of what kinds of knowledge these models learn during pretraining . Reformulating tasks as fill-in-the-blanks problems (e.g., cloze tests) is a natural approach to gauging such knowledge . Using AutoPrompt, we show that maskedlanguage models (MLMs) have an inherent capability to perform sentimentanalysis and natural language inference without additional parameters orfinetuning .…

We present a method for adversarial attack detection based on the inspection of a sparse set of neurons . We follow the hypothesis that adversarial attacksintroduce imperceptible perturbations in the input and that these perturbationchange the state of neurons relevant for the concepts modelled by the attackedmodel .…

The aim of the recommender system is to provide personalized suggestions to users, not to suggest popular items . This results in the terrible Matthew effect, making popular items be more frequently recommended and become even more popular . To eliminate popularity bias, it is essential to answer the question that what the ranking score would be if the model onlyuses item property .…

We present a regression method for enhancing the predictive power of a model by exploiting expert knowledge in the form of shape constraints . The method is conceptually simple, applicable in multiple dimensions . It is tested andvalidated on two real-world manufacturing processes, namely laser glass bendingand press hardening of sheet metal .…

We propose using federated learning, a decentralized on-device learning paradigm, to train speech recognition models . By performing epochs of training on a per-user basis, we must incur the cost of dealing with non-IID data distributions, which are expected to negatively affect the quality of the trained model .…

We introduce a Bayesian (deep) model-based reinforcement learning method(RoMBRL) that can capture model uncertainty to achieve sample-efficient policyoptimisation . We show that RoMBRL outperforms existing approaches on many challenging control benchmark tasks in terms of sample complexity and task performance .…

We consider off-policy evaluation (OPE) in continuous action domains . In OPE, one aims to learn thevalue under a new policy using historical data generated by a differentbehavior policy . We develop a brand-new deep jumpQ-evaluation method for OPE . The key ingredient of our method lies inadaptively discretizing the action space using deep jump Q-learning.…

Deep reinforcement learning (RL) algorithms are powerful tools for solving motor motor decision tasks . However, the trained models are often difficult tointerpret, because they are represented as end-to-end deep neural networks . In this paper, we shed light on the inner workings of such trained models by analyzing the pixels that they attend to during task execution .…

Activelearning refers to the development of algorithms to judiciously pick limitedsubsets of unlabeled samples that can be sent for labeling by an oracle . When obtaining labels is expensive, the requirement of a large labeled training data set for deep learning can be mitigated by active learning .…

We establish fundamental upper bounds on the amount of secret key that can be extracted from continuous variable quantum Gaussian states by using only localGaussian operations, local classical processing, and public communication . For one-way communication, we prove that the key is bounded by the R\’enyi-$2$Gaussian entanglement of formation $E_{F,2}^{\mathrm{\scriptscriptstyle G}}$ The same is true if two-way public communication is allowed but Alice and Bob employ protocolsthat start with destructive local Gaussian measurements .…

Honey Encryption is an approach to encrypt the messages using low min-entropykeys, such as weak passwords, OTPs, PINs, credit card numbers . But the currenttechniques used in producing the decoy plaintexts do not model human languageentirely . A gibberish, random assortment of words is not enough to fool anattacker; that will not be acceptable and convincing, whether or not the attacker knows some information of the genuine source .…

Producing or sharing Child Sexual Exploitation Material (CSEM) is a serious crime fought vigorously by Law Enforcement Agencies . When an LEA seizes a computer from a potential producer or consumer of CSEM, they need to analyze the hard disk’s files looking for pieces of evidence .…

We empirically analyze a simple heuristic for large sparse set coverproblems . It uses a weighted greedy algorithm as a basic building block . By multiplying updates of the weights attached to the elements, the greedysolution is improved . The implementation of this algorithm is trivial and the algorithm is essentially free of parameters that would requiretuning .…

In this note, we revisit the recursive random contraction algorithm of Kargerand Stein for finding a minimum cut in a graph . We show that the analysis becomes particularly clean in the analysis of graphs . We also consider other similar variants of the algorithm, and show that no such algorithm can achieve an asymptotically better probability of finding a fixed minimum cut .…

Quantum algorithms for both differential equation solving and for machinelearning potentially offer an exponential speedup over all known classical algorithms . The essential obstacle for quantumdifferential equation solving is that outputting useful information may requiredifficult post-processing, and for quantum machinelearning is that inputting the training set is a difficult task just by itself .…

Hub labeling schemes are popular methods for computing distances on roadnetworks and other large complex networks, often answering to a query within microseconds for graphs with millions of edges . However, things take a different turn when the hub labels have a sublogarithmicsize .…

Dimensionality reduction is a crucial first step for many unsupervised learning tasks including anomaly detection . We propose to use a minimum spanning tree (MST),a graph-based algorithm, to approximate the local neighborhood structure andgenerate structure-preserving distances among data points . We find that using the MST regularizer improves the performance of anomaly detection substantially forboth generative adversarial networks .…

Emotions recognition is commonly employed for health assessment . However, thetypical metric for evaluation in therapy is based on patient-doctor appraisal . Machine learning algorithms can be a useful tool for the classification ofemotions . We propose a tool which enables users to take speech samples and identify a range of emotions from audio elements through a machine learning model .…

Decoding sequences that stem from multiple transmissions of a codeword over an insertion, deletion, and substitution channel is a critical component of efficient deoxyribonucleic acid (DNA) data storage systems . We propose two new algorithms for inference from multiple received sequences .…

We introduce a novel convolutional network architecture with an interpretablelatent space for modeling guitar amplifiers . The proposed system intuitivelycombines or subtracts characteristics of different amplifiers, allowing musicians to design entirely new guitar timbres .…

Papaya2 is a software to calculate 2D higher-order shape metrics with a libraryinterface, support for Irreducible Minkowski Tensors and interpolated marchingsquares . Extensions to Matlab, JavaScript and Python are provided as well . We are not aware of other open-source software which provides higher-rank shape characterization in2D .…

We study two-sided decentralized matching markets in which participants have uncertain preferences . We present a statistical model to learn the preferences of participants . We derive an optimal strategy that maximizes the agent’sexpected payoff and calibrate the uncertain state by taking the opportunitycosts into account .…

AutoAtlas consists of two neural network components: one that performs multi-label partitioning based on local texture . A second that compresses the information contained withineach partition . We train both of these components simultaneously by optimizing a loss function that is designed to promote accurate reconstruction of eachpartition .…

We draw connections between simple neural networks and under-determinedlinear systems to comprehensively explore several interesting theoreticalquestions in the study of neural networks . We emphatically show that itis unsurprising such networks can achieve zero training loss . Our lower bounds grow more slowly with data set size than existing work that trains the hidden layer weights .…

Pulse-agile radar systems have demonstrated favorable performance in dynamicelectromagnetic scenarios . The use of non-identical waveforms within aradar’s coherent processing interval may lead to harmful distortion effect . This paper presents an online learningframework to optimize detection performance while mitigating harmful sidelobelevels .…

This paper studies an ES sharing model where multiple buildings cooperativelyinvest and share a community ES (CES) to harness economic benefits from on-siterenewable integration and utility price arbitrage . We propose a fair cost allocation based on nucleolus by employing aconstraints generation technique .…

Anomaly detection is concerned with identifying data patterns that deviateremarkably from the expected behaviour . This is an important research problem, due to its broad set of application domains, from data analysis to e-health,cybersecurity, predictive maintenance, fault prevention, and industrialautomation .…

We prove that given a discrete space with $n$ points which is either embeddedin a system of $k$ trees, or the Cartesian product of trees, we can compute all eccentricities in ${\cal O}(2^ . O’O”(N+n)^{1+o(1)$ time . This is nearoptimal under the Strong Exponential-Time Hypothesis, even in the very specialcase of an $n .-vertex…

Speech separation has been well-developed while there are still problemswaiting to be solved . The main problem we focus on in this paper is the frequent label permutation switching of permutation invariant training (PIT) For N-speaker separation, there would be N!…

Conventional NN-based methods use twobranches for a sound event detection (SED) target and a direction-of-arrival (DOA) target . Using two networks dedicated to each task increases system complexity and network size . We propose an activity-coupled Cartesian DOA (ACCDOA)representation . The ACCDOA representation enables us to solve a SELD task with a single target and has two advantages: avoiding thenecessity of balancing the objectives and model size increase .…

The proposed system yields state-of-the-art performance on content-based music recommendation tested with millions of users and tracks . The results show the generalization ability of our audio embeddings. The results are the results of the proposed system.Also, we extract audio embeddeddings as features for music genre classificationtasks.…

Software comprehension and maintenance activities, such as refactoring, are said to be negatively impacted by software complexity . Most teams using a plugin forrefactoring (JDeodorant) reduced software complexity more effectively and withsimpler processes than those using only Eclipsenative features . We were able to find moderate correlations (43%) between software cyclomatic complexity and process cyclomatic complexity .…

Data stalls are a major overhead in main-memory database engines due to the use of pointer-rich data structures . Lightweight coroutines ease the implementation of software prefetching to hide data stalls . CoroBase can perform 2x better for read-intensive workloads and remain competitive for those workloads that do not benefit from software pre-etching .…

Software-controlled accounts (i.e., bots) are one of the main actors associated with manipulationcampaigns . Uncovering the strategies behind bots’ activities is of paramount importance to detect and curb such campaigns . We present a long term (one year) analysis of bots activity on Twitter in the run-up to the 2018 U.S.…

IQIYI voice conversion system (T24) for VoiceConversion 2020 . In the competition, each target speaker has 70 sentences . The evaluation results show that this system canachieve better voice conversion effects. In the case of using 16k rather than 24k sampling rate audio, the conversion result is relatively good innaturalness and similarity.…

A new multifactormodel for the election result prediction based on Twitter data has be developed for this purpose . The model was tested by attempting to predict the results of the US 2020 elections in November . The parameters for 3 November were calculated as -0.213423 for Democratsand 0.0455818 for Republicans .…

Botnets may be used to betterinfiltrate the social graph over time and to create an illusion of community behavior, amplifying their message and increasing persuasion . We analyzed a dense crawl of a subset of Twitter traffic amounting to nearly all interactions by Greek-speaking Twitter users for a period of 36 months .…

An hieroglyph on n letters is a cyclic sequence of the letters 1,2, . 2n such that each letter appears in the sequence twice . We give a criterion for weak realizability of a disk with ribbons corresponding to H can be cut out of the M\”obius strip .…

Following the success of Spotify, the role of theagile coach has branched out in terms of tasks and responsibilities, but little research has been conducted to examine how this role is practiced . The most essential traits of an agilecoach are being emphatic, people-oriented, able to listen, diplomatic, and persistent .…