Yiwei Zhao

Yiwei Zhao

Auto-generated profile, based on publicly available information

Yiwei Zhao

Carnegie Mellon University, Pittsburgh, USA

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 June 2023
Verified
Conference Paper
Open access
HKHongbo Kang
Hongbo Kang
YZYiwei Zhao
Yiwei Zhao
GBGuy E. Blelloch
Guy E. Blelloch
LDLaxman Dhulipala
Laxman Dhulipala
YGYan Gu
Yan Gu
PGPhillip B. Gibbons
Phillip B. Gibbons
Memory latency and bandwidth are significant bottlenecks in designing in-memory indexes. Processing-in-memory (PIM), an emerging hardware design approach, alleviates this problem by embedding processors in memory modules, enabling low-latency memory access whose aggregated bandwidth scales linearly with the number of PIM modules. Despite recent work in balanced comparison-based indexes on PIM systems, building efficient tries for PIMs remains an open challenge due to tries' inherently unbalanced shape. This paper presents the PIM-trie, the first batch-parallel radix-based index for PIM systems that provides load balance and low communication under adversary-controlled workloads. We introduce trie matching-matching a query trie of a batch against the compressed data trie-as a key building block for PIM-friendly index operations. Our algorithm combines (i) hash-based comparisons for coarse-grained work distribution/elimination and (ii) bit-by-bit comparisons for fine-grained matching. Combined with other techniques (meta-block decomposition, selective recursive replication, differentiated verification), PIM-trie supports LongestCommonPrefix, Insert, and Delete in O(logP) communication rounds per batch and O(l/w) communication volume per string, where P is the number of PIM modules, l is the string length in bits, and w is the machine word size. Moreover, work and communication are load-balanced among modules whp, even under worst-case skew.

18 July 2023
Verified
Conference Paper
Open access
HKHongbo Kang
Hongbo Kang
YZYiwei Zhao
Yiwei Zhao
GBGuy E. Blelloch
Guy E. Blelloch
LDLaxman Dhulipala
Laxman Dhulipala
YGYan Gu
Yan Gu
PGPhillip B. Gibbons
Phillip B. Gibbons
Processing-in-memory (PIM) is an emerging technology to alleviate the high cost of data movement by pushing computation into/near memory modules. There is an inherent tension, however, between minimizing communication (data movement) and achieving load balance in PIM systems in the presence of workload skew. This work introduces PIM-tree, a PIM-based index that simultaneously achieves low communication, good load balance, and low space consumption. It achieves good theoretical bounds in the PIM Model and efficient on a real-world PIM machine, outperforming prior PIM-based and state-of-the-art CPU-based indexes.

16 July 2025
Verified
Conference Paper
Open access
YZYiwei Zhao
Yiwei Zhao
HKHongbo Kang
Hongbo Kang
YGYan Gu
Yan Gu
GBGuy E. Blelloch
Guy E. Blelloch
LDLaxman Dhulipala
Laxman Dhulipala
PGPhillip B. Gibbons
Phillip B. Gibbons

20 January 2025
Verified
Conference Paper
Open access
YZYiwei Zhao
Yiwei Zhao
JCJinhui Chen
Jinhui Chen
SZSai Qian Zhang
Sai Qian Zhang
SSSyed Shakib Sarwar
Syed Shakib Sarwar
KSKleber Hugo Stangherlin
Kleber Hugo Stangherlin
ZLZiyun Li
Ziyun Li
Low-latency and low-power edge AI is crucial for Augmented/Virtual Reality applications. Recent advances demonstrate that hybrid models, combining convolution layers (CNN) and transformers (ViT), often achieve a superior accuracy/performance tradeoff on various computer vision and machine learning (ML) tasks. However, hybrid ML models can present system challenges for latency and energy efficiency due to their diverse nature in dataflow and memory access patterns. In this work, we leverage architecture heterogeneity from Neural Processing Units (NPU) and Compute-In-Memory (CIM) and explore diverse execution schemas for efficient hybrid model executions. We introduce H4H-NAS, a two-stage Neural Architecture Search (NAS) framework to automate the design of hybrid CNN/ViT models for heterogeneous edge systems featuring both NPU and CIM. We propose a two-phase incremental supernet training in our NAS to resolve gradient conflicts between sampled subnets caused by different block types in a hybrid model search space. Our H4H-NAS approach is also powered by a performance estimator built with NPU performance results measured on real silicon, and CIM performance based on industry IPs. H4H-NAS searches hybrid CNN-ViT models with fine granularity and achieves significant (up to 1.34%) top-1 accuracy improvement on ImageNet-1k. Moreover, results from our algorithm/hardware co-design reveal up to 56.08% overall latency and 41.72% energy improvements by introducing heterogeneous computing over baseline solutions. Overall, our framework guides the design of hybrid network architectures and system architectures for NPU+CIM heterogeneous systems.

28 January 2026
Verified
Conference Paper
Open access
YZYiwei Zhao
Yiwei Zhao
HKHongbo Kang
Hongbo Kang
ZMZiyang Men
Ziyang Men
YGYan Gu
Yan Gu
GBGuy E. Blelloch
Guy E. Blelloch
PGPhillip B. Gibbons
Phillip B. Gibbons