Error submitting Object Detection job to ai-platform

I am using the instructions laid out in the documentation to submit a job to Google Clouds ai-platform. My pipeline runs correctly both locally and when deploying to a Google Cloud VM.

The command I am running is below, it is practically the same as the documentation, except updated for the latest runtime versions. I have also tried the GPU training instructions with the same result

cp object_detection/packages/tf2/setup.py .
gcloud ai-platform jobs submit training `whoami`_object_detection_`date +%m_%d_%Y_%H_%M_%S` \
    --job-dir=${MODEL_DIR} \
    --package-path ./object_detection \
    --module-name object_detection.model_main_tf2 \
    --runtime-version 2.3 \
    --python-version 3.7 \
    --scale-tier BASIC_TPU \
    --region us-central1 \
    -- \
    --use_tpu true \
    --model_dir=${MODEL_DIR} \
    --pipeline_config_path=${PIPELINE_CONFIG_PATH}

In the AI Platform job details I am receiving the error message below. I have seen this error brought up in a couple of other places, including here. But nothing is referenced, which it would appear I have the ability to configure in the context of submitting a job to the ai-platform

The replica master 0 exited with a non-zero status of 1. 
Traceback (most recent call last):
  File "/opt/conda/lib/python3.7/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/opt/conda/lib/python3.7/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/root/.local/lib/python3.7/site-packages/object_detection/model_main_tf2.py", line 31, in <module>
    import tensorflow.compat.v2 as tf
  File "<frozen importlib._bootstrap>", line 983, in _find_and_load
  File "<frozen importlib._bootstrap>", line 967, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 668, in _load_unlocked
  File "<frozen importlib._bootstrap>", line 638, in _load_backward_compatible
  File "/opt/conda/lib/python3.7/site-packages/wrapt/importer.py", line 156, in load_module
    module = self.loader.load_module(fullname)
  File "/root/.local/lib/python3.7/site-packages/tensorflow/__init__.py", line 436, in <module>
    _ll.load_library(_main_dir)
  File "/root/.local/lib/python3.7/site-packages/tensorflow/python/framework/load_library.py", line 153, in load_library
    py_tf.TF_LoadLibrary(lib)
tensorflow.python.framework.errors_impl.NotFoundError: /opt/conda/lib/python3.7/site-packages/tensorflow/core/kernels/libtfkernel_sobol_op.so: undefined symbol: _ZN10tensorflow8OpKernel11TraceStringEPNS_15OpKernelContextEb