rights reserved. Presenter: Haichen Shen, Yao Wang Amazon SageMaker Neo, Deep Engine Science Dynamic Model in TVM AWS AI© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Models with models© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Support dynamic model in TVM ● Support Any-dim in typing ● Use shape function to compute the type at runtime ● Virtual input_name = "data" input_shape = [tvm.relay.Any(), 3, 224, 224] dtype = "float32" block = get_model('resnet50_v1', pretrained=True) mod, params = relay.frontend.from_mxnet(block, shape={input_name:

Open Discussion on Project Planning Planning in an Agile Environment Key Tenets  Planning should focus strongly on the near term  Eliminate waste caused by planning for eventualities that never come

Change Happening Faster Than Ever? Yes, It Is • AI User + Usage + CapEx Growth = Unprecedented • AI Model Compute Costs High / Rising + Inference Costs Per Token Falling = Performance Converging + Developer • AI Usage + Cost + Loss Growth = Unprecedented • AI Monetization Threats = Rising Competition + Open-Source Momentum + China’s Rise • AI & Physical World Ramps = Fast + Data-Driven • Global Internet on Page 975 …Charts Paint Thousands of Words… AI Monetization Threats = Rising Competition + Open-Source Momentum + China’s Rise 5 Leading USA LLMs vs. China LLM Desktop User Share Note: Data

lexicon for this guide. Later sections will present more detail on the Agile processes and their application in DoD programs. The foundational structure of an Agile program is: Release - Capability system or commercial off-the-shelf (COTS) product, or building a small-scale or self-constrained application. In other words, Agile works well when the program needs to modify software for government purposes team or the target end user cannot be accessed. Program scope is mostly limited to the application layer while using existing infrastructure. Program Scope Program spans core capabilities and

. . . . . . . . . . . . . . . . . . . . . . . . . 1624 101.26 Fixing precompilation hangs due to open tasks or IO . . . . . . . . . . . . . . . . . 1626 102 Developing/debugging Julia’s C code 1630 combinations of argument types, and applied by dis- patching to the most specific matching definition. This model is a good fit for mathematical programming, where it is unnatural for the first argument to "own" guage. In addition to the above, some advantages of Julia over comparable systems include: • Free and open source (MIT licensed) • User-defined types are as fast and compact as built-ins • No need to vectorize

. . . . . . . . . . . . . . . . . . . . . . . . . 1624 101.26 Fixing precompilation hangs due to open tasks or IO . . . . . . . . . . . . . . . . . 1626 102 Developing/debugging Julia’s C code 1630 combinations of argument types, and applied by dis- patching to the most specific matching definition. This model is a good fit for mathematical programming, where it is unnatural for the first argument to "own" guage. In addition to the above, some advantages of Julia over comparable systems include: • Free and open source (MIT licensed) • User-defined types are as fast and compact as built-ins • No need to vectorize

. . . . . . . . . . . . . . . . . . . . . . . . . 1929 105.30 Fixing precompilation hangs due to open tasks or IO . . . . . . . . . . . . . . . . . 1934 106 Developing/debugging Julia's C code 1938 combinations of argument types, and applied by dis- patching to the most specific matching definition. This model is a good fit for mathematical programming, where it is unnatural for the first argument to "own" guage. In addition to the above, some advantages of Julia over comparable systems include: • Free and open source (MIT licensed) • User-defined types are as fast and compact as built-ins • No need to vectorize

. . . . . . . . . . . . . . . . . . . . . . . . . 1929 105.30 Fixing precompilation hangs due to open tasks or IO . . . . . . . . . . . . . . . . . 1934 106 Developing/debugging Julia's C code 1938 combinations of argument types, and applied by dis- patching to the most specific matching definition. This model is a good fit for mathematical programming, where it is unnatural for the first argument to "own" guage. In addition to the above, some advantages of Julia over comparable systems include: • Free and open source (MIT licensed) • User-defined types are as fast and compact as built-ins • No need to vectorize

. . . . . . . . . . . . . . . . . . . . . . . . . 1929 105.30 Fixing precompilation hangs due to open tasks or IO . . . . . . . . . . . . . . . . . 1934 106 Developing/debugging Julia's C code 1938 combinations of argument types, and applied by dis- patching to the most specific matching definition. This model is a good fit for mathematical programming, where it is unnatural for the first argument to "own" guage. In addition to the above, some advantages of Julia over comparable systems include: • Free and open source (MIT licensed) • User-defined types are as fast and compact as built-ins • No need to vectorize

. . . . . . . . . . . . . . . . . . . . . . . . . 1982 106.31 Fixing precompilation hangs due to open tasks or IO . . . . . . . . . . . . . . . . . 1988 107 Developing/debugging Julia's C code 1991 combinations of argument types, and applied by dis- patching to the most specific matching definition. This model is a good fit for mathematical programming, where it is unnatural for the first argument to "own" guage. In addition to the above, some advantages of Julia over comparable systems include: • Free and open source (MIT licensed) • User-defined types are as fast and compact as built-ins • No need to vectorize