22 April 2020 in IEEE Intelligent Systems
Machine learning relies on the availability of vast amounts of data for training. However, in reality, data are mostly scattered across different organizations and cannot be easily integrated due to many legal and practical constraints. To address this important challenge in the field of machine learning, we introduce a new technique and framework, known as federated transfer learning (FTL), to improve statistical modeling under a data federation. FTL allows knowledge to be shared without compromising user privacy and enables complementary knowledge to be transferred across domains in a data federation, thereby enabling a target-domain party to build flexible and effective models by leveraging rich labels from a source domain. This framework requires minimal modifications to the existing model structure and provides the same level of accuracy as the non-privacy-preserving transfer learning. It is flexible and can be effectively adapted to various secure multi-party machine learning tasks.
10 July 2018 in IEEE Transactions on Information Theory
Like classical block codes, a locally repairable code also obeys the Singleton-type bound (we call a locally repairable code optimal if it achieves the Singleton-type bound). In the breakthrough work of Tamo and Barg, several classes of optimal locally repairable codes were constructed via subcodes of Reed-Solomon codes. Thus, the lengths of the codes given by Tamo and Barg are upper bounded by the code alphabet size q. Recently, it was proved through the extension of construction by Tamo and Barg that the length of q-ary optimal locally repairable codes can be q +1 by Jin et al. Surprisingly, Barg et al. presented a few examples of q-ary optimal locally repairable codes of small distance and locality with code length achieving roughly q 2 . Very recently, it was further shown in the work of Li et al. that there exist q-ary optimal locally repairable codes with the length bigger than q+1 and the distance proportional to n. Thus, it becomes an interesting and challenging problem to construct new families of q-ary optimal locally repairable codes of length bigger than q+1. In this paper, we construct a class of optimal locally repairable codes of distances 3 and 4 with unbounded length (i.e., length of the codes is independent of the code alphabet size). Our technique is through cyclic codes with particular generator and parity-check polynomials that are carefully chosen.
28 December 2016 in IEEE Transactions on Information Theory
Both Maximum Distance Separable and Euclidean self-dual codes have theoretical and practical importance and the study of MDS self-dual codes has attracted lots of attention in recent years. In particular, determining the existence of q-ary MDS self-dual codes for various lengths has been investigated extensively. The problem is completely solved for the case where q is even. This paper focuses on the case where q is odd. We construct a few classes of new MDS self-dual codes through generalized Reed-Solomon codes. More precisely, we show that for any given even length n, we have a q-ary MDS code as long as q ≡ 1 mod 4 and q is sufficiently large (say q ≥ 4 n × n 2 ). Furthermore, we prove that there exists a q-ary MDS self-dual code of length n if q = r 2 and n satisfies one of the three conditions: 1) n ≤ r and n is even; 2) q is odd and n - 1 is an odd divisor of q - 1; and 3) r ≡ 3 mod 4 and n=2tr for any t ≤ (r - 1)/2.
7 June 2018 in IEEE Transactions on Information Theory
Constructing locally repairable codes achieving Singleton-type bound (we call them optimal codes in this paper) is a challenging task and has attracted great attention in the last few years. Tamo and Barg [14] first gave a breakthrough result in this topic by cleverly considering subcodes of Reed-Solomon codes. Thus, q-ary optimal locally repairable codes from subcodes of Reed-Solomon codes given in [14] have length upper bounded by q. Recently, it was shown through extension of construction in [14] that length of q-ary optimal locally repairable codes can be q+1 in [7]. Surprisingly it was shown in [2] that, unlike classical MDS codes, q-ary optimal locally repairable codes could have length bigger than q + 1. Thus, it becomes an interesting and challenging problem to construct q- ary optimal locally repairable codes of length bigger than q + 1. In the present paper, we make use of rich algebraic structures of elliptic curves to construct a family of q-ary optimal locally repairable codes of length up to q + 2 √ q. It turns out that locality of our codes can be as big as 23 and distance can be linear in length.
10 January 2019 in IEEE Transactions on Information Theory
A locally repairable code (LRC) with locality r allows for the recovery of any erased codeword symbol using only r other codeword symbols. A Singleton-type bound dictates the best possible tradeoff between the dimension and distance of LRCs-an LRC attaining this tradeoff is deemed optimal. Such optimal LRCs have been constructed over alphabets growing linearly in the block length. Unlike the classical Singleton bound, however, it was not known if such a linear growth in the alphabet size is necessary or, for that matter, even if the alphabet needs to grow at all with the block length. Indeed, for small code distances 3 and 4, arbitrarily long optimal LRCs were known over fixed alphabets. Here, we prove that for distances d ≥ 5, the code length n of an optimal LRC over an alphabet of size q must be at most roughly O(dq 3 ). For the case d = 5, our upper bound is O(q 2 ). We complement these bounds by showing the existence of optimal LRCs of length Ω d,r (q 1+1/〈(d-3)/2〉 ) when d r + 2. These bounds match when d = 5, thus pinning down n = Θ(q 2 ) as the asymptotically largest length of an optimal LRC for this case.
10 March 2021 in IEEE Transactions on Information Theory
Insertion and deletion (insdel for short) errors are synchronization errors in communication systems caused by the loss of positional information in the message. Reed-Solomon codes have gained a lot of interest due to its encoding simplicity, well structuredness and list-decoding capability in the classical setting. This interest also translates to the insdel metric setting, as the Guruswami-Sudan decoding algorithm can be utilized to provide a deletion correcting algorithm in the insdel metric. Nevertheless, there have been few studies on the insdel error-correcting capability of Reed-Solomon codes. Our main contributions in this article are explicit constructions of two families of 2-dimensional Reed-Solomon codes with insdel error-correcting capabilities asymptotically reaching those provided by the Singleton bound. The first construction gives a family of Reed-Solomon codes with insdel error-correcting capability asymptotic to its length. The second construction provides a family of Reed-Solomon codes with an exact insdel error-correcting capability up to its length. Both our constructions improve the previously known construction of 2-dimensional Reed-Solomon codes whose insdel error-correcting capability is only logarithmic on the code length.